Management processes for insulin therapy using reinforcement learning and therapy escalation pathways

Abstract

A method for escalating therapy in diabetic patients includes receiving user glucose data from an in vivo glucose monitoring device, receiving first therapy information for a first therapy, the first therapy including basal insulin, calculating one or more glucose metrics based on the received glucose data, titrating the dose of basal insulin based on one or more glucose metrics, and determining basal excess based on the glucose data and one or more of the first therapy information. Advantageously, the system can periodically monitor the user's glucose adjustments, detect basal excess, provide frequent therapy interventions and adjustments, reduce the duration of intervention, and increase user adherence, outcomes, and satisfaction.

Description

Technical Field

[0001] The present disclosure relates to therapy management devices, systems, and methods, such as for detecting the effectiveness of current therapies, detecting basal over-parameterization, escalating to one or more therapy pathways, reinforcement learning, and management processes for insulin therapy using software application devices, systems, and methods.

Background Art

[0002] The detection and / or monitoring of blood glucose levels (glucose values) can be extremely important for the health of people suffering from diabetes. Persons with diabetes (PWDs) generally need to monitor their blood glucose levels to ensure that the blood glucose levels are maintained within clinically safe ranges, and further use this information to determine whether and / or when insulin is needed to lower the blood glucose levels in the body, and / or when additional glucose is needed to raise the blood glucose levels in the body. Many systems, such as continuous glucose monitors (CGMs) for measuring interstitial fluid glucose levels, enable individuals to monitor their glucose. Some of these systems include electrochemical biosensors, which use glucose sensors adapted to be placed in vivo, for example, with complete or partial insertion into a subcutaneous or transdermal site in the body, for continuous in vivo monitoring of blood glucose levels from the body fluid (e.g., interstitial fluid) at that site.

[0003] Diabetes mellitus is a chronic metabolic disorder caused by the inability of the human pancreas to produce sufficient amounts of the hormone insulin, resulting in impaired metabolism that does not allow for proper absorption of sugars and starches. This dysfunction can lead to hyperglycemia, such as the presence of excessive amounts of glucose. Persistent hyperglycemia is associated with a variety of serious symptoms and long-term life-threatening complications, such as dehydration, ketoacidosis, diabetic coma, cardiovascular disease, chronic renal failure, retinal damage, and nerve damage with an risk of amputation.

[0004] To maintain blood glucose levels within acceptable limits, a permanent treatment that provides consistent blood glucose control is necessary. Such blood glucose control can be achieved by lowering elevated blood glucose levels through the regular external supply of medication to the body of the patient with PWD. Externally administered bioeffective drugs (e.g., insulin or its analogues) are generally given by daily injections. In some cases, multiple daily injections of a mixture of rapid-acting (RA) insulin and long-acting (LA) insulin are administered via a reusable transdermal fluid delivery device.

[0005] Glycated hemoglobin (HbA1c, or A1c) is a form of hemoglobin that chemically binds to sugar. The formation of sugar-hemoglobin bonds can indicate the presence of excess sugar in the bloodstream, which often indicates diabetes when present at high concentrations. A1c is primarily measured to determine the 3-month average blood glucose level and can be used as a diagnostic test for diabetes mellitus and as an assessment test for blood glucose control in prediabetes disease (PWD). Generally, a normal A1c level is less than 5.7%, levels between 5.7% and 6.4% indicate prediabetes, and levels above 6.5% indicate diabetes. Within the prediabetic range of 5.7% to 6.4%, higher A1c levels in PWD indicate a higher risk of developing type 2 diabetes.

[0006] Currently, monitoring A1C levels is limited to every 3 to 6 months, depending on various factors. This method of measuring patient progression and follow-up, limited to quarterly or semi-annual visits, can contribute to significant therapeutic inertia throughout PWD treatment, resulting in suboptimal outcomes and a "treat-to-failure" approach to diabetes therapy. Furthermore, limited human resources continue to perpetuate delays in adhering to treatment standards. [Overview of the project] [Problems that the invention aims to solve]

[0007] Therefore, there is a need to develop a therapy management system that can regularly monitor users' blood glucose control, provide frequent (e.g., daily or weekly, rather than every 3-6 months) therapeutic interventions and adjustments, reduce the duration of interventions, and increase user adherence, outcomes, and satisfaction. [Means for solving the problem]

[0008] In one embodiment, a non-temporary computer-readable storage medium stores instructions, which, when executed by a processor, cause the processor to perform operations for a management process for insulin therapy, the operations comprising: performing a survey of a patient with diabetes (PWD); receiving survey information in response to the survey; and performing a high-touch dialogue phase with respect to the PWD, the high-touch dialogue phase comprising iterative adjustments to insulin therapy based on the collected data and published guidelines for insulin therapy, the high-touch dialogue phase establishing at least basal dose settings, meal dose settings, and corrective dose settings for the PWD; and performing a progress phase with respect to the PWD after the completion of the high-touch dialogue phase, the progress phase comprising: collecting performance metrics for the PWD; applying reinforcement learning to the performance metrics; and automatically performing actions with respect to the PWD based on the reinforcement learning.

[0009] The implementation may include one or all of the following features: A high-touch interaction phase is performed over a predetermined period of time. The high-touch interaction phase has a goal-based duration. The goal-based duration depends on the determination that the PWD's blood glucose level is within the target range and that the PWD is being administered insulin as specified in insulin therapy. The operation further includes evaluating each of several insights and determining whether the insight has been triggered, taking into account at least performance metrics and collected data, and the action is selected based on whether at least one of the insights has been triggered. Insights are evaluated according to their priority relative to the PWD. The insight to be evaluated is selected from the set of insights based on which has the highest priority. Evaluating at least one of the insights includes multimodal evaluation. The operation further includes performing observations on the PWD's state after automatically performing the action. Reinforcement learning includes providing positive or negative feedback on the action selection. The operation further includes performing a cooldown after the observation has finished. The behavior further includes evaluating each of several insights and determining whether the insight has been triggered, taking into account at least performance metrics and collected data, and the action is selected based on whether at least one of the insights has been triggered. The behavior further includes checking whether a cooldown is associated with the action before selecting an action. If a cooldown is associated with the action, the behavior evaluates the next insight from the several insights instead. The behavior further includes evaluating each of several insights and determining whether the insight has been triggered, taking into account at least performance metrics and collected data, and the action is selected based on whether at least one of the insights has been triggered. The behavior further includes checking whether an observation is associated with an insight before selecting an action.If the observation leads to an insight, the action evaluates improvement criteria for the insight, taking into account at least performance metrics and collected data. Evaluating improvement criteria includes multimodal evaluation. For any insights not triggered in the evaluation, the action further includes determining whether the insight has not been triggered for at least a predetermined time and taking streak actions based on that determination. Automating the action includes selecting one or more goals for the PWD and presenting one or more goals to the PWD. The action further includes receiving opt-in or opt-out input from the PWD.

[0010] In some embodiments, a method for therapeutic escalation for PWD may include receiving user glucose data from an in vivo glucose monitoring device. In some embodiments, the method may further include receiving first therapeutic information for a first therapeutic. In some embodiments, the first therapeutic may include basal insulin. In some embodiments, the method may further include calculating one or more glucose metrics based on the received glucose data. In some embodiments, the method may further include titrating the dose of basal insulin based on one or more glucose metrics. In some embodiments, the method may further include determining basal excess based on glucose data and one or more of the first therapeutic information.

[0011] In some embodiments, the method may further include outputting a recommendation to add a second therapy if basal excess is determined. In some embodiments, outputting a recommendation may include displaying a recommendation. In some embodiments, displaying a recommendation may include displaying the recommendation on a display device, a remote device (e.g., a receiver, smartphone, computer, etc.), a drug delivery device, a pen cap, or a combination thereof.

[0012] In some embodiments, outputting a recommendation may include outputting a recommended therapy. In some embodiments, outputting a recommended therapy may include communicating the recommended therapy to a display device, a remote device (e.g., a receiver, smartphone, computer, etc.), a drug delivery device, a pen cap, or a combination thereof.

[0013] In some embodiments, the method may further include administering a recommended dose and / or recommended therapy by a drug delivery device. In some embodiments, the administration of the recommended dose and / or recommended therapy may be automated or at user confirmation.

[0014] In some embodiments, the first therapeutic information may include information regarding a recommended dose of basal insulin, an actual dose of basal insulin, the timing of basal insulin administration, or a combination thereof.

[0015] In some embodiments, the method may further include outputting an indicator that titration has been stopped if basal excess has been detected, and / or an indicator that the patient should consult a physician if basal excess has been detected, rather than a recommendation to add a second therapy. For example, this indicator could provide the user with a notification (e.g., "Titration has been stopped because basal insulin could not be optimized. Please consult your doctor," or "Please consult your doctor regarding basal excess.").

[0016] In some embodiments, one or more glucose metrics may include mean glucose, median glucose, glucose time-in-range (TIR), glucose time-below-range (TBR), glucose time-above-range (TAR), glucose time-very-low (TVL), glucose management index (GMI), minimum wake glucose (MMG), minimum post-administration glucose (MPDG), postprandial glucose (PPG), sleep-to-wake glucose (BeAM), time-in-tight-range (TITR), time-in-very-tight-range (TIVTR), or a combination thereof.

[0017] In some embodiments, MMG can be configured to estimate fasting blood glucose using continuous glucose monitoring (CGM) data. In some embodiments, MMG can be calculated based on the lowest morning glucose value, for example, between approximately 4:30 AM and 10:00 AM, at least about 30 minutes before the estimated wake-up time. In some embodiments, MMG can be calculated based on the median MMG with the number of days involving nocturnal activity (e.g., treatment for hypoglycemia, intake of RA dose, absence of LA dose, etc.) excluded from the calculation, thereby providing a more robust estimate of nocturnal glucose. In some embodiments, MMG can be used as a useful metric for identifying patients who require adjustment of LA insulin therapy (e.g., MMG 14 days prior to the day to increase and / or decrease the LA insulin therapy setting). In some embodiments, MMG can be used as a sensitive indicator of nocturnal hypoglycemia compared to more traditional fasting glucose estimates (e.g., pre-breakfast glucose). In some embodiments, a change in median MMG may indicate a change in LA insulin therapy (e.g., increase or decrease). In some embodiments, an increase in median MMG (e.g., from 110 mg / dL to 155 mg / dL) may indicate an increase in LA insulin therapy. In some embodiments, a decrease in median MMG (e.g., from 155 mg / dL to 110 mg / dL) may indicate a decrease in LA insulin therapy.

[0018] In some embodiments, the first therapeutic information can be manually entered via input to a computing device. In some embodiments, the first therapeutic information can be collected by a drug delivery device and communicated to a computing device. In some embodiments, the first therapeutic information can be collected by a dose detection device for the drug delivery device, such as an add-on module or smart pen cap, among other things. In some embodiments, the drug delivery device may include, among other things, an infusion pump, a patch pump, or an injection pen. In some embodiments, the initial dose can also be calculated from data collected from one or more smart scales, other glucose measuring devices, or a combination thereof.

[0019] In some embodiments, basal excess can be determined based on the ratio of the total daily dose (TDD) of insulin to the user's body weight. In some embodiments, basal excess can be determined based on glucose variability. In some embodiments, basal excess can be determined based on hypoglycemia metrics. In some embodiments, basal excess can be determined based on changes in blood glucose levels during a meal or overnight.

[0020] In some embodiments, the method may further include reducing the basal insulin dose when a second therapy is added. In some embodiments, the second therapy may include a glucagon-like peptide-1 (GLP-1) receptor agonist. In some embodiments, the method may further include dose titration of the GLP-1 receptor agonist. In some embodiments, dose titration may include receiving user input regarding side effects of the second therapy. In some embodiments, dose titration may include recommending a dose increase if the user input is associated with no side effects. In some embodiments, dose titration may include receiving the user's body weight measurement. In some embodiments, dose titration may include recommending a dose increase if the body weight measurement exceeds the target body weight.

[0021] In some embodiments, one or more glucose metrics may include postprandial glucose elevation for each of several meals. In some embodiments, the method may further include recommending the initiation of a planar insulin dose for the first meal if the postprandial glucose elevation for one meal exceeds a threshold. In some embodiments, the threshold may be defined by the total change in blood glucose levels from the pre-meal period to the post-meal period (difference between minimum and maximum values), the rate of change in blood glucose levels (slope), the difference between the post-planar and pre-planar doses, or a combination thereof. In some embodiments, the method may further include recommending the initiation of a planar insulin dose for several meals if the postprandial glucose elevation for each of several meals exceeds a threshold. In some embodiments, the threshold may be defined by the total change in blood glucose levels from the pre-meal period to the post-meal period for each of several meals (difference between minimum and maximum values), the rate of change in blood glucose levels for each of several meals (slope), the difference between the post-planar and pre-planar doses for each of several meals, or a combination thereof.

[0022] In some embodiments, a therapy escalation system for diabetic patients may include an in vivo glucose monitoring device, a remote device, a drug delivery device, and a processor. In some embodiments, the in vivo glucose monitoring device may be configured to measure the user's glucose data. In some embodiments, the remote device may communicate with the in vivo glucose monitoring device. In some embodiments, the remote device may be configured to receive or retrieve glucose data from the in vivo glucose monitoring device. In some embodiments, the drug delivery device may communicate with the in vivo glucose monitoring device and the remote device. In some embodiments, the drug delivery device may be configured to administer one or more dosing regimens. In some embodiments, the processor may communicate with the sample measurement system, the remote device, and the drug delivery device. In some embodiments, the processor may be coupled to a memory that stores instructions causing the processor to perform an action, if executed, which includes receiving the user's glucose data from the in vivo glucose monitoring device. In some embodiments, the action may further include receiving first therapy information for a first therapy from the remote device, the drug delivery device, or both. In some embodiments, the first therapy may include basal insulin. In some embodiments, the operation may further include calculating one or more glucose metrics based on received glucose data. In some embodiments, the operation may further include titrating the dose of basal insulin based on one or more glucose metrics. In some embodiments, the operation may further include determining basal excess based on glucose data and one or more of the first therapeutic information.

[0023] In some embodiments, the operation may further include outputting a recommendation to the remote device to add a second therapy if a basal excess is detected. In some embodiments, outputting a recommendation can include displaying the recommendation on a remote device (e.g., a receiver, smartphone, computer, etc.). In some embodiments, displaying the recommendation can include displaying the recommendation on a display device, remote device, drug delivery device, pen cap, or a combination thereof.

[0024] In some embodiments, outputting a recommendation can include outputting a recommended therapy. In some embodiments, outputting a recommended therapy can include communicating the recommended therapy to a display device, remote device (e.g., a receiver, smartphone, computer, etc.), drug delivery device, pen cap, or a combination thereof.

[0025] In some embodiments, the operation can further include administering a recommended dose and / or recommended therapy by a drug delivery device. In some embodiments, administering a recommended dose and / or recommended therapy can be done automatically or upon user confirmation.

[0026] In some embodiments, a system for managing a therapy to maintain one or more user metrics at or within one or more targets can include a specimen measurement system, a remote device, a drug delivery device, and a software application. In some embodiments, the specimen measurement system can be configured to measure a user's specimen (e.g., glucose). In some embodiments, the specimen measurement system can include a specimen sensor. In some embodiments, the remote device can communicate with the specimen measurement system. In some embodiments, the remote device can be configured to receive or retrieve sensor data from the specimen sensor. In some embodiments, the drug delivery device can communicate with the specimen measurement system and the remote device. In some embodiments, the drug delivery device can be configured to administer one or more dosing regimens. In some embodiments, the software application can communicate with the specimen measurement system, the remote device, and the drug delivery device.

[0027] In some embodiments, a software application can run on a processor coupled to a memory storing instructions that, when executed, cause the processor to perform operations including recommending a first dosing regimen (e.g., basal insulin) when one or more user metrics are outside of one or more targets. In some embodiments, the operations can further include determining a basal excess of the first dosing regimen. In some embodiments, the operations can further include adding a second dosing regimen (e.g., GLP-1, single post-meal dose, post-meal dose for each meal) when one or more user metrics remain outside of one or more targets or when a basal excess is determined. In some embodiments, the operations can further include determining a basal excess of the second dosing regimen. In some embodiments, a user can receive an initial dosing regimen (e.g., GLP-1, etc.) prior to the first dosing regimen (e.g., basal insulin).

[0028] In some embodiments, one or more user metrics can include mean glucose, central glucose, glucose time in range (TIR), glucose time in tight range (TITR), glucose time in very tight range (TIVTR), glucose time in below range (TBR), glucose time in above range (TAR), glucose time in valley low (TVL), glucose coefficient of variation (CV), glucose management indicator (GMI), minimum morning glucose (MMG), minimum post-dose glucose (MPDG), postprandial glucose (PPG), basal bolus ratio (BBR), bedtime to wakeup glucose (BeAM), insulin sensitivity factor (ISF), postprandial rapid-acting with correction glucose (PRA-CG), weight, BMI, total daily dose (TDD) (e.g., of insulin), ratio of TDD to weight, A1C level (e.g., percentage of glycated hemoglobin, mmol / mol), heart rate, blood pressure, or a combination thereof.

[0029] In some embodiments, one or more targets may include a target blood glucose level (e.g., approximately 110 mg / dL), a target blood glucose range (e.g., approximately 110 mg / dL to approximately 155 mg / dL), a target basal state (e.g., MMG approximately 100 mg / dL to approximately 130 mg / dL), a target diet state (e.g., MPDG approximately 100 mg / dL to approximately 120 mg / dL), a target modified ISF state (e.g., PRA-CG approximately 70 mg / dL to approximately 180 mg / dL), a target body weight, a target A1C level (e.g., 5.7%), a target heart rate, a target blood pressure, a target body mass, or a combination thereof. In some embodiments, one or more targets may be based at least in part on the user's body mass index (BMI) and the user's mean blood glucose level.

[0030] In some embodiments, the operation in the second dosing regimen may further include monitoring one or more user compliance metrics over a first period. In some embodiments, one or more user compliance metrics may include proper wear and use of a CGM, appropriate and consistent administration of the drug, or a combination thereof. In some embodiments, the operation in the second dosing regimen may further include rapid titration of a basal analog, a GLP-1 class drug, or mealtime RA insulin over a second period. In some embodiments, rapid titration may include recommending an adjusted dose at a first interval (e.g., daily) based on one or more user metrics at that first interval until no change is observed over a predetermined period (e.g., 3 days). In some embodiments, the operation in the second dosing regimen may further include monitoring for no change in dose output status over a third period. In some embodiments, the operation in the second dosing regimen may further include maintaining the titration of a basal analog, a GLP-1 class drug, or mealtime RA insulin over a fourth period. In some aspects, maintaining titration may include recommending an adjusted dose over a second interval (e.g., several days, one week) based on one or more user metrics for that second interval, until a new issue (e.g., a new medication regimen, a new exercise routine, a new diet, etc.) is identified.

[0031] In some embodiments, the software application can be configured to determine basal excess based at least in part on drug delivery information. In some embodiments, basal excess may be determined at least in part on the ratio of insulin TDD to the user's body weight that exceeds a predetermined threshold. In some embodiments, for example, the predetermined threshold may be about 0.5 U units (U) / kg / day. In some embodiments, the software application can be configured to determine basal excess based at least in part on glucose metrics. In some embodiments, basal excess may be determined at least in part on the increase in the difference between waking (AM) and bedtime (PM) and / or the difference between post-prandial and pre-prandial. In some embodiments, the software application can be configured to determine basal excess based at least in part on hypoglycemia metrics that do not exceed a predetermined threshold. In some embodiments, for example, the predetermined threshold may be a glucose concentration of about 70 mg / dL. In some embodiments, the software application can be configured to determine basal excess based at least in part on glucose variability that exceeds a predetermined threshold. In some embodiments, for example, a predetermined threshold may include a pre-meal rise of less than approximately 30 mg / dL, a post-meal spike of less than approximately 110 mg / dL, the total change in blood glucose levels from the pre-meal period to the post-meal period (the difference between the minimum and maximum values), or a combination thereof. In some embodiments, the software application may be configured to determine basal excess based at least in part on a predetermined threshold (e.g., a basal insulin dose exceeding 0.5 U / kg / day), an increase in the ratio of insulin TDD to the user's body weight, an increase in the morning-to-bedtime (AM)-to-nighttime (PM) difference and / or the post-to-pre-to-prandial difference, a hypoglycemic metric exceeding a predetermined threshold (e.g., a glucose concentration of approximately 70 mg / dL), an increase in the variability of sensor data, or a combination thereof.

[0032] In some embodiments, the operation may further include recommending the addition of a second dosing regimen of a glucagon-like peptide-1 (GLP-1) receptor agonist or a dual gastric suppressant peptide (GIP) / GLP-1 receptor agonist if one or more user metrics remain outside one or more targets. In some embodiments, the operation may further include recommending the addition of a second dosing regimen of a GLP-1 receptor agonist or a dual GIP / GLP-1 receptor agonist if basal excess is determined.

[0033] In some embodiments, the operation may further include recommending the addition of a third dosing regimen of prandial insulin per meal if one or more user metrics remain outside one or more targets. In some embodiments, the operation may further include recommending the addition of a third dosing regimen of prandial insulin per meal if basal excess is determined.

[0034] In some embodiments, the operation may further include recommending the addition of a fourth dosing regimen of multiple daily injections (MDIs) of prandial insulin if one or more user metrics remain outside one or more targets. In some embodiments, the operation may further include recommending the addition of a fourth dosing regimen of MDIs of prandial insulin if basal excess is determined.

[0035] In some embodiments, the operation may further include recommending the addition of a fifth dosing regimen of basal insulin and prandial insulin for each meal if one or more user metrics remain outside one or more targets. In some embodiments, the operation may further include recommending the addition of a fifth dosing regimen of basal insulin and postprandial (booster) insulin for each meal if basal excess is determined.

[0036] In some embodiments, for any of the dosing regimens described above, the operation may further include performing a monitoring phase, a rapid titration phase, and a maintenance phase for the current dosing regimen. In some embodiments, the monitoring phase may include monitoring one or more user compliance metrics of the current dosing regimen over a predetermined period (e.g., 3 days) to determine whether the user is properly wearing and using the CGM and whether the medication is being administered appropriately and consistently. In some embodiments, one or more user compliance metrics may include the frequency of glucose data (scans), gaps in glucose data, timing of administration, dose size (e.g., corresponding to a recommended dose), receipt of user input, or a combination thereof. In some embodiments, once the user compliance metrics are met, the rapid titration phase may include recommending an adjusted dose at a first interval (e.g., daily at midnight) based on one or more user metrics of that first interval until no changes are observed over a predetermined period (e.g., 3 days). In some embodiments, dose adjustments may be made at regular intervals (e.g., an increase of 0.25 U, a 10% increase) or proportional to the degree of glycemic dysfunction (e.g., larger dose changes if the patient is significantly deviating from the glucose target). In some embodiments, the maintenance titration phase may include recommending adjusted doses at a second interval (e.g., several days, one week) based on one or more user metrics for that second interval, until new issues (e.g., a new medication regimen, a new exercise routine, a new diet, etc.) are identified.

[0037] In some embodiments, a software application can be configured to adjust the dose of one or more dosing regimens based at least in part on one or more metrics of the user. In some embodiments, one or more metrics may include mean glucose, glucose TIR, glucose TBR, glucose TVL, GMI, MMG, PPG, BBR, BeAM, ISF, or a combination thereof.

[0038] In some embodiments, the system may further include an external scale that communicates with a software application. In some embodiments, the external scale may be configured to measure the user's weight. In some embodiments, the external scale may be configured to provide the user's weight to the software application to determine basal excess, at least in part, based on the ratio of insulin TDD to the user's measured weight, which exceeds a predetermined threshold (e.g., a basal insulin dose exceeding 0.5 U / kg / day). In some embodiments, a remote device may prompt the user to input a weight measurement or to weigh themselves using the external scale.

[0039] In some embodiments, the system may further include external sensors that communicate with a software application. In some embodiments, the external sensors may be configured to measure one or more user parameters. In some embodiments, one or more user parameters may include dietary parameters, exercise-related parameters, activity parameters, respiratory parameters, heart rate, heart rate variability, body temperature, blood pressure, sleep-related parameters, nausea parameters, or a combination thereof. In some embodiments, the external sensors may include accelerometers, gyroscopes, microelectromechanical systems (MEMS) devices, position sensors, voice sensors, smart devices, smartphones, smart rings, bed sensors, smart pill bottles, smart injection pen caps, or a combination thereof.

[0040] In some embodiments, the operation may further include receiving input signals from a user or a third party. In some embodiments, the operation may further include sending control signals to a software application to recommend adjustments to one or more dosing regimens based at least in part on the input signals. In some embodiments, the input signals may include feedback, clinician guidelines, user adherence, verbal instructions, therapy routes, therapy escalations, therapy sessions, therapy session analysis history, or a combination thereof. In some embodiments, the third party may include a medical professional, such as a clinician, primary care physician (PCP), or health care professional (HCP), among others.

[0041] In some embodiments, the operation may further include performing analysis of sensor data and user data to generate a predictive model. In some embodiments, the predictive model may be based on regression, model-based parameter fitting, supervised machine learning, unsupervised machine learning, neural networks, classification models, or a combination thereof.

[0042] In some embodiments, the software application can be configured to receive one or more user parameters measured over time from a remote server. In some embodiments, the software application can be configured to recommend adjustments to one or more dosing regimens based at least in part on an algorithm or function incorporating one or more user parameters measured over time. In some embodiments, the software application can be configured to receive a constructed user profile. In some embodiments, the constructed user profile can be generated using an artificial intelligence (AI) module trained using analysis from a remote server. In some embodiments, the software application can be configured to recommend adjustments to one or more dosing regimens based at least in part on the constructed user profile.

[0043] In some embodiments, the processor may be part of a sample measurement system, a remote device, or a drug delivery device. In some embodiments, the software application may be part of an external device, a remote server, or a cloud server. In some embodiments, the software application (e.g., one or more algorithms) may be executed by a combination of devices.

[0044] In some embodiments, communication between the sample measurement system and the remote device may include wireless communication, near-field communication (NFC), Bluetooth, Bluetooth Low Energy (BLE), or a combination thereof.

[0045] In some embodiments, a therapy escalation method for diabetic patients to detect non-adherence to basal insulin recommendations may include receiving user glucose data from an in vivo glucose monitoring device. In some embodiments, the method may further include recommending a first dose of basal insulin to the user. In some embodiments, the method may further include calculating one or more glucose metrics based on the received glucose data. In some embodiments, one or more glucose metrics may include a minimum glucose metric. In some embodiments, the method may further include titrating a recommended dose of basal insulin based on one or more glucose metrics. In some embodiments, the method may further include recommending a second dose of basal insulin to the user. In some embodiments, the second dose may differ from the first dose, for example, due to titration of the recommended dose. In some embodiments, the method may further include calculating the change in the minimum glucose metric from a first period before recommending the second dose to a second period after recommending the second dose. In some embodiments, the method may further include determining whether the change in the minimum glucose metric is outside a predetermined metric. In some embodiments, the method may further include outputting an indicator that basal insulin titration has stopped if the change in the minimum glucose metric is outside a predetermined metric.

[0046] In some embodiments, the minimum glucose metric may include fasting glucose, minimum daily glucose (DMG), minimum morning glucose (MMG), average glucose over a given period, minimum hourly average glucose (DMHAG), or a combination thereof. In some embodiments, for example, the minimum glucose metric may include DMHAG.

[0047] In some embodiments, a given metric may include the change in the minimum glucose metric being less than a given percentage of the expected change in the minimum glucose metric. In some embodiments, for example, the change in the minimum glucose metric may be less than 50% of the expected change in the minimum glucose metric. In some embodiments, the expected change in the minimum glucose metric is based on one or more user metrics. In some embodiments, one or more user metrics may include body weight, BMI, age, insulin sensitivity factor (ISF), total daily dose (TDD), TDD-to-body weight ratio, A1C level, heart rate, blood pressure, or a combination thereof. In some embodiments, for example, one or more user metrics may include ISF. In some embodiments, the expected change in the minimum glucose metric may be based on the user's glucose data history. In some embodiments, the expected change in the minimum glucose metric may be proportional to the change between a first dose and a second dose.

[0048] In some embodiments, a given metric may include a minimum glucose metric change being less than a predetermined change in blood glucose level. In some embodiments, a given metric may include a minimum glucose metric change being less than 5 mg / dL. In some embodiments, a given metric may include a minimum glucose metric change being less than 15 mg / dL. In some embodiments, a given metric may include a minimum glucose metric change being less than 10 mg / dL. In some embodiments, a given metric may include a minimum glucose metric change being less than 3 mg / dL. In some embodiments, a given metric may include a minimum glucose metric change being less than 1 mg / dL. In some embodiments, a given metric may include a minimum glucose metric change being less than 0.1 mg / dL. In some embodiments, a given metric may include a minimum glucose metric change being less than the range of approximately 0.1 mg / dL to approximately 15 mg / dL.

[0049] In some embodiments, a given metric may include determining a change in the minimum glucose metric based on a statistical method. In some embodiments, a given metric may include a statistical test of the minimum glucose metric that does not support the hypothesis that the minimum glucose metric has changed. In some embodiments, the statistical test may include an insulin sensitivity test (IST), an insulin tolerance test (ITT), an oral glucose tolerance test (OGTT), a fasting plasma glucose (FPG) test, a random plasma glucose test, or a combination thereof.

[0050] In some embodiments, the method may further include prompting the user to confirm the administration of a second dose. In some embodiments, the method may further include restarting titration if the user has confirmed the administration of a second dose. In some embodiments, the method may further include recommending a third dose based on the titration of the second dose if the user has not confirmed the administration of a second dose.

[0051] In some embodiments, the method may further include recommending limiting dose changes of basal insulin if, after titrating the recommended dose over a predetermined period, there is no change in the minimum glucose metric. In some embodiments, limiting dose changes may include setting an upper limit so that the dose change does not exceed a predetermined percentage of the previous dose, for example, 1%, 2%, 5%, 10%, etc. In some embodiments, limiting dose changes may include stopping any dose changes. In some embodiments, the predetermined period may include at least 3 days.

[0052] In some embodiments, the method may further include stopping the upward titration of basal insulin if, after titrating the recommended dose over a predetermined period, the minimum glucose metric does not decrease. In some embodiments, the method may further include stopping the downward titration of basal insulin if, after titrating the recommended dose over a predetermined period, the minimum glucose metric does not increase.

[0053] In some embodiments, the method may further include activating a blind mode so that glucose data is not displayed to the user. In some embodiments, the method may further include starting a counter configured to monitor the number of titration cycles if the change in the minimum glucose metric is outside a predetermined metric. In some embodiments, the method may further include stopping basal insulin titration if, after a predetermined number of titration cycles in which the dose of basal insulin has been increased or decreased, the change in the minimum glucose metric remains outside the predetermined metric.

[0054] In some embodiments, the method may further include receiving basal insulin dose administration time data. In some embodiments, titration of the recommended dose is not based on basal insulin dose data.

[0055] In some embodiments, the method may further include outputting a prompt to the user to confirm that the user is following the recommended dosage. In some embodiments, the method may further include outputting a prompt to the user to ask the user for guidance from a healthcare professional. In some embodiments, the method may further include outputting a quiz to the user to verify that the user is following the recommended dosage.

[0056] In some embodiments, a method for detecting non-adherence to a dose recommendation may include receiving the user's glucose data for a first period from an in vivo glucose monitoring device. In some embodiments, the method may further include determining a first minimum glucose metric for the first period. In some embodiments, the method may further include providing the user with a dose recommendation based on the glucose data for the first period. In some embodiments, the method may further include receiving the user's glucose data for a second period following the dose recommendation. In some embodiments, the method may further include determining a second minimum glucose metric for the second period. In some embodiments, the method may further include determining non-adherence to a dose recommendation based on a comparison of the first minimum glucose metric and the second minimum glucose metric. In some embodiments, the method may further include outputting an index of non-adherence.

[0057] In some embodiments, the first minimum glucose metric and the second minimum glucose metric may include the minimum hourly average glucose (DMHAG) of the day. In some embodiments, the comparison may include a change between a first minimum glucose metric and a second minimum glucose metric. In some embodiments, determining non-adherence may include comparing the change between the first minimum glucose metric and the second minimum glucose metric to a predetermined percentage change threshold. In some embodiments, determining non-adherence may include comparing the change between the first minimum glucose metric and the second minimum glucose metric to a predetermined change in blood glucose level.

[0058] In some embodiments, the comparison may include a direction between a first minimum glucose metric and a second minimum glucose metric. In some embodiments, the direction is inversely proportional to the dose recommendation. In some embodiments, for example, user non-adherence can be determined by a direction (e.g., a trend) in which the first and second minimum glucose metrics increase (do not decrease) in response to an increase in the dose recommendation. In some embodiments, for example, user non-adherence can be determined by a direction (e.g., a trend) in which the first and second minimum glucose metrics decrease (do not increase) in response to a decrease in the dose recommendation.

[0059] Any implementation of any of the technologies described above may include a system, method, process, device, and / or apparatus. Details of one or more implementations are described in the accompanying drawings and the following description. Other features will become apparent from the specification and drawings, as well as the claims.

[0060] Further features and exemplary embodiments of this disclosure, as well as the structure and operation of various embodiments, are described in detail below with reference to the accompanying drawings. It should be noted that embodiments are not limited to those described herein. Such embodiments are presented herein for illustrative purposes only. Further embodiments will become apparent to those skilled in the art based on the teachings contained herein.

[0061] The accompanying drawings incorporated herein and forming part thereof illustrate embodiments and, together with the description herein, explain the principles of the embodiments and further enable those skilled in the art to fabricate and use the embodiments. [Brief explanation of the drawing]

[0062] [Figure 1] This is an example of a process that guides PWDs when managing their own insulin therapy. [Figure 2] An example of a pre-training survey for the process shown in Figure 1 is presented. [Figure 3]Figure 1 shows an example of a patient messaging profile that can be constructed using the process shown. [Figure 4] Here is an example of a system for managing diabetes. [Figure 5] This shows an example of a system for guiding PWD (Periodic Warning Day) when managing insulin therapy. [Figure 6] Figure 5 shows an example of a state diagram that can be used by the insight generation engine. [Figure 7] Figure 5 schematically shows the insight prioritization that can be performed by the insight generation engine. [Figure 8A] Figure 5 shows an example of a process that can be performed by the insight generation engine. [Figure 8B] Figure 5 shows an example of a process that can be performed by the insight generation engine. [Figure 9] An example of multimodal in-sight triggering is shown. [Figure 10] Here is an example of a multimodal insight satisfaction evaluation. [Figure 11] Here is an example of an action related to insights. [Figure 12] Examples of computing devices that can be used to implement the technologies described herein are shown. [Figure 13] This is a schematic diagram of a therapeutic management system in an exemplary manner. [Figure 14A] This is a schematic control diagram of a therapeutic escalation pathway in an exemplary manner. [Figure 14B] This is a schematic therapeutic assessment of a therapeutic escalation pathway, based on an exemplary embodiment. [Figure 14C] This is a schematic control diagram of a therapeutic escalation pathway in an exemplary manner. [Figure 15] This is a schematic control diagram of a first dosing regimen in an exemplary embodiment. [Figure 16] This is a schematic control diagram for a second dosing regimen, in an exemplary embodiment. [Figure 17] This is a schematic control diagram for a third dosing regimen, in an exemplary embodiment. [Figure 18] This is a schematic control diagram for a fourth dosing regimen, in an exemplary embodiment. [Figure 19A] This is a schematic diagram of the basal state according to an exemplary embodiment. [Figure 19B] This is a schematic diagram of a mealtime, illustrating an exemplary configuration. [Figure 19C] This is a schematic diagram of the corrected insulin sensitivity factor (ISF) status according to an exemplary embodiment. [Figure 20] A flowchart illustrating therapies escalation for diabetic patients is shown in an exemplary manner. [Figure 21] A flowchart illustrating an exemplary embodiment for detecting non-adherence to dose recommendations is shown. [Figure 22] A flowchart illustrating an exemplary embodiment for detecting non-adherence to dose recommendations is shown. [Figure 23] This is a schematic diagram of a computing device that can be used to implement the operations described herein, in an exemplary embodiment. [Modes for carrying out the invention]

[0063] The features and exemplary aspects of this disclosure will become more apparent when the detailed descriptions described below are read in conjunction with the drawings, in which similar reference numerals identify corresponding elements throughout. In the drawings, similar reference numerals generally indicate identical, functionally similar, and / or structurally similar elements. In addition, generally, the leftmost digit of the reference numeral identifies the drawing in which the reference numeral first appears. Unless otherwise indicated, the drawings provided throughout this disclosure should not be construed as to scale.

[0064] This specification describes examples of systems and techniques that use reinforcement learning as a method for applying machine learning techniques to manage insulin therapy for patients with impaired dysregulation (PWD) and thereby gaining benefits from them. Generally, reinforcement learning can be applied to aspects of the insulin therapy management process in which the algorithm does not have pre-specified actions to perform and therefore can consider indirect feedback when planning those actions. The systems according to this disclosure are superior in communicating with PWD throughout the course of insulin therapy because they take into account the PWD's current physiological and psychological presence over a wide range of factors (sometimes referred to as taking into account the PWD's state perfectly).

[0065] Provided herein are systems, apparatus, devices, methods, and / or computer program product embodiments, and / or combinations and partial combinations thereof, for therapeutic escalation pathways for managing diabetes and maintaining euglycemic function, such as by maintaining one or more glucose metrics at or within a target level or range.

[0066] The system described below can continuously monitor a user's blood glucose levels, determine one or more glucose metrics, implement one or more dosing regimens, determine user adherence to therapy, determine whether basal excess exists for therapy, monitor therapy over discrete periods, and adjust and / or escalate therapy based on one or more measured user metrics in accordance with one or more dosing regimens.

[0067] This specification discloses one or more embodiments incorporating the features of this disclosure. The embodiments described, and references in this specification such as “one embodiment,” “an embodiment,” “an exemplary embodiment,” and “an exemplary embodiment,” indicate that the embodiments described may include certain features, structures, or characteristics, but not all embodiments may include such features, structures, or characteristics. Furthermore, such phrases do not necessarily refer to the same embodiments. Moreover, if certain features, structures, or characteristics are described in relation to one embodiment, it will be understood that achieving such features, structures, or characteristics in relation to other embodiments, whether explicitly described or not, is within the scope of the knowledge of those skilled in the art.

[0068] As used herein, the terms “about,” “substantially,” or “approximately” indicate a value of a given quantity that may vary based on a particular technique. Based on a particular technique, the terms “about,” “substantially,” or “approximately” may indicate a value of a given quantity that varies within, for example, 1 to 15% of the value (e.g., ±1%, ±2%, ±5%, ±10%, or ±15% of the value). In some embodiments, the terms “substantially” and “about” as used herein are used to describe and explain small fluctuations, such as those resulting from variations in processing. For example, these terms may refer to ±5% or less, e.g., ±2% or less, e.g., ±1% or less, e.g., ±0.5% or less, e.g., ±0.2% or less, e.g., ±0.1% or less, e.g., ±0.05% or less. Also, as used herein, indefinite articles such as “a” or “an” mean “at least one.”

[0069] Numerical values, including the endpoints of the range, may be expressed herein as approximations preceded by terms such as “about,” “substantially,” or “approximately.” In such cases, other aspects include specific numerical values. Whether numerical values ​​are expressed as approximations or not, this disclosure includes two aspects: one in which they are expressed as approximations and the other in which they are not. It will be further understood that one endpoint of each range is significant in relation to and independently of another endpoint.

[0070] Aspects of the Disclosure may be implemented in hardware, firmware, software, or any combination thereof. Aspects of the Disclosure may also be implemented as instructions stored in a machine-readable medium that can be read and executed by one or more processors. The machine-readable medium may include any mechanism for storing or transmitting information in a format readable by a machine (e.g., a computing device). For example, the machine-readable medium may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media, optical storage media, flash memory devices, propagating signals in electrical, optical, acoustic or other forms (e.g., carrier waves, infrared signals, digital signals, etc.), and others. Furthermore, firmware, software, routines, and / or instructions may be described herein as performing certain actions. However, such descriptions are merely for convenience, and it should be understood that such actions actually result from computing devices, processors, controllers, or other devices that perform firmware, software, routines, instructions, etc.

[0071] As used herein, the term “overbasalization” refers to titration of basal insulin beyond an appropriate dose in an attempt to achieve a blood glucose target. In some embodiments, the appropriate dose is the dose that allows the user to reach the blood glucose target.

[0072] As used herein, the terms “long-acting (LA) insulin” and “basal insulin” are interchangeable terms and refer to the dose of insulin associated with non-meal times and overnight.

[0073] As used herein, the terms “rapid-acting (RA) insulin,” “prandial insulin,” and “bolus insulin” are interchangeable terms and refer to the amount of insulin administered in relation to a meal.

[0074] As used herein, the term “therapy intervention” may mean initiating or modifying a therapy regimen over a period of time (e.g., adjusting the dose, adjusting the type of dose), or recommending the initiation or modification of therapy (e.g., daily and / or weekly adjustments).

[0075] The term "duration of intervention" can refer to the time between therapeutic interventions. For example, the intervention period may span a predetermined period (e.g., one day, two days, one week, two weeks, etc.).

[0076] The term "monitoring phase" may refer to monitoring one or more user compliance metrics of the current therapy over a predetermined period (e.g., one day, two days, three days, etc.).

[0077] As used herein, the term “titrating a dose” may mean determining to maintain or adjust (e.g., increase or decrease) a dose in order to improve glucose control.

[0078] As used herein, the term “upward titration” refers to an increase in dose. As used herein, the term “downward titration” refers to reducing the dose.

[0079] As used herein, the term “titration cycle” may refer to a cycle in which a first dose is recommended, one or more glucose metrics are calculated, and the first dose administered is maintained or adjusted based on one or more glucose metrics.

[0080] As used herein, the terms “add a second therapy” or “adding a second therapy” may refer to a drug therapy regimen comprising one or more therapies, and it is possible to add or incorporate a second therapy into a drug therapy regimen, resulting in a drug therapy regimen comprising a first therapy and a second therapy (e.g., basal insulin therapy and GLP-1 therapy, basal insulin therapy and prandial insulin therapy, etc.).

[0081] As used herein, the term “inversely proportional” may refer to two quantities that behave such that a decrease in a second quantity leads to an increase in a first quantity, or vice versa. For example, an increase in dose recommendation leads to a decrease in the direction (e.g., trend) of the first and second minimum glucose metrics, or vice versa.

[0082] Exemplary management process for insulin therapy As described above, reinforcement learning can be applied to a form of insulin therapy management process in which the algorithm does not have pre-specified actions to perform, and therefore can consider indirect feedback when planning those actions. The system according to this disclosure is adept at communicating with PWD throughout the course of insulin therapy because it takes into account the PWD's current physiological and psychological presence over a wide range of factors (sometimes referred to as taking into account the PWD's state perfectly).

[0083] In some implementations, a reinforcement learning-driven insight engine can be provided that seeks to optimally guide PWDs through their insulin injection treatment journey. The insight engine can use PWD preferences, healthcare provider instructions, physiological data, behavioral data, drug administration data, and outcome measures for PWDs and other similar PWDs to select coaching messages and therapy adjustments. For example, coaching messages can instruct PWDs on how changes in their behavior can benefit their diabetes treatment journey. Another example of therapy adjustments may include any changes to insulin therapy to improve glucose performance (e.g., changing the type of insulin, changing the timing of insulin intake, changing the mental model regarding the administration of the type of insulin (e.g., long-acting insulin), changing the insulin dose (e.g., long-acting or rapid-acting insulin), or changing the correction table for insulin). The main goal of the insight engine may be to understand each patient and determine the best way to help them achieve control of their diabetes. With the goal of encountering PWDs at some point in their treatment journey, whether the PWD initiates a new therapy, optimizes an existing one, or promotes recommendations for therapy changes, the system can leverage various methods to assess the PWD's current state and provide clinically relevant recommendations. The insight engine can use generalized reinforcement learning to guide the PWD toward better therapy, including but not limited to insulin settings, blood glucose measurement, injection administration, and / or injection timing control. System components may include, but are not limited to, blood glucose meters or monitors, insulin pumps, smart caps on injection pens, insulin therapy applications, and / or cloud infrastructure.

[0084] This disclosure may provide a management process for insulin therapy that includes an initial interaction with the patient (PWD), questioning the PWD and / or healthcare provider to obtain preferences, training the PWD, subjecting the PWD to a high-touch interaction phase in which insulin settings for the PWD can be rapidly adjusted based on standard guidelines, and finally conducting a progression phase aimed at optimizing insulin therapy using reinforcement learning. The high-touch interaction phase and one or more other initial phases may provide preparation for subsequent optimization. In such a management process, the system or technique may consider a holistic view of the PWD's general state, as opposed to mere specific measurements. From such a state, the system / technique may probabilistically determine the next action to take (e.g., increase, decrease, or maintain the dose, and / or make another recommendation or perform coaching). After performing the action, the system / technique may observe one or more aspects of the PWD to evaluate whether the PWD took the correct action, thereby obtaining feedback as part of reinforcement learning. In the future, if a similar situation occurs / has occurred, the system / technique may consider such feedback. For example, feedback can increase (or decrease) the likelihood that the same action will be repeated.

[0085] Insulin delivery devices include, but are not limited to, insulin pens, insulin inhalers, insulin pumps, and insulin syringes. Inappropriate insulin administration is always a concern, whether due to human error, insulin pen malfunction, skipped doses, double dosing, or inaccurate administration. The methods, devices, and systems provided herein are described for insulin delivery, glucose data collection, and / or treatment of diabetes. Furthermore, the methods, devices, and systems provided herein may be adapted for the delivery of other drugs, the collection of other specimen data, and / or treatment of other diseases. The methods, devices, and systems provided herein are described by illustrating the features and functionality of many exemplary embodiments. Other implementations are also possible.

[0086] Some examples in this specification refer to insulin injection pens. An insulin injection pen includes at least one container (e.g., an insulin cartridge) for holding insulin, a dial or other mechanism for specifying a dose, and a pen needle for transdermal delivery of insulin into the tissues or vascular structures of a diabetic patient. In reusable insulin injection pens, the insulin container (e.g., a cartridge) is replaceable or refillable. Prefilled insulin injection pens are intended for use for a limited time. The dose setting mechanism may include a rotatable wheel coupled to the mechanism and / or electronic equipment for setting an upper limit on the amount of insulin to be administered in a specified dose volume (e.g., in terms of units of insulin). The dose setting mechanism may have a mechanical and / or electronic display that reflects the current setting of the mechanism. The pen needle may be permanently attached to the housing of the insulin injection pen (e.g., as in the case of a disposable pen) or removable (e.g., so that a new needle can be applied as needed). For example, a replaceable pen needle may include a hollow needle fixed to a mounting component configured to be removablely attached to the end of the insulin injection pen.

[0087] Some examples in this specification refer to mobile communication devices. As used herein, mobile communication devices include, but are not limited to, mobile phones, smartphones, wearable electronic devices (e.g., smartwatches), tablets, laptop computers, portable computers, and similar devices. A mobile communication device includes one or more processors, non-temporary storage devices (e.g., memory and / or hard drives) that hold executable instructions for operating the mobile communication device, wireless communication components, and one or more input and / or output devices (e.g., touchscreens, displays, or keyboards). A mobile communication device may operate according to one or more application programs stored locally on the mobile communication device, stored remotely (e.g., when using cloud computing), or a combination thereof. A mobile communication device may run at least one operating system to perform functions and services.

[0088] Some examples in this specification refer to blood glucose meters (BGMs). A BGM is an electronic device configured to receive a sample (e.g., a blood sample) from a patient with diabetes (PWD) and analyze the sample in order to estimate the blood glucose level of a diabetic patient. A BGM may be configured to have a new test strip partially inserted within the BGM for each sample, and the patient then places a drop of blood on the end of the test strip extending from the BGM. The BGM performs a test on that drop of blood as it passes through the test strip.

[0089] Some examples in this specification refer to continuous glucose monitors (CGMs). A CGM is an electronic device configured to continuously or periodically read blood glucose measurements to estimate the blood glucose level of a patient with impaired vision (PWD). A CGM may have subcutaneously placed electrodes that transmit their output to a receiver, such as a handheld device, (e.g., via a transmitter). Another type of CGM may be fully implanted subcutaneously in the PWD. Yet another type of CGM may be non-invasive, avoiding skin puncture, for example, by analyzing the PWD's breath or by shining light onto the PWD's skin. Some CGMs periodically determine blood glucose levels (e.g., after seconds or minutes) and output the information automatically or at a prompted time. A CGM may include one or more wireless communication components for signaling, including but not limited to near-field communication (NFC) and / or Bluetooth communication.

[0090] In some embodiments, the systems, devices, and / or methods provided herein can recommend an insulin dose (e.g., dose of long-acting insulin and / or rapid-acting insulin) using any preferred technology. In some embodiments, the recommended insulin dose may be based on blood glucose data (e.g., current estimated blood glucose (EGV), blood glucose trend data, etc. from a continuous glucose monitor, flash glucose monitor, blood glucose meter, or any other sensor), insulin administration data (calculated values ​​such as bolus dose of rapid-acting insulin, dose of long-acting insulin, administration time, insulin onboard (IOB), and / or active insulin), meal data (meal time, user-estimated carbohydrates, user-estimated meal classification, user-estimated meal glucose effect, user meal history, user meal trends, etc.), and / or one or more insulin delivery parameters (e.g., total daily dose of basal insulin or long-acting insulin, carbohydrate-to-insulin ratio (CR), insulin sensitivity factor (ISF), etc.). In some embodiments, the methods, devices, and systems provided herein can adjust insulin delivery parameters over time based on glucose data and / or insulin administration data.

[0091] In this specification, some examples refer to long-acting insulin and rapid-acting insulin, or, in some cases, more generally, to the first and second types of insulin. Insulin used for therapeutic treatment is often synthetic human insulin. Furthermore, various insulins can be characterized by the typical speed at which they begin to work in the body of a diabetic patient after administration, and / or the typical length of time they remain active in the body of a diabetic patient. Rapid-acting insulin can be used for administration with meals or to correct hyperglycemia. There are two or more types of insulin that can be considered rapid-acting insulin. Many, though not all, rapid-acting insulins begin to work within about an hour after administration. Similarly, there are two or more types of insulin that can be considered long-acting insulin, and many, though not all, long-acting insulins begin to work about an hour or more after administration. Long-acting insulin is often called basal insulin (e.g., insulin used to support basal metabolic needs). Generally, long-acting insulin has a longer duration of activity (i.e., the length of time insulin remains active in a diabetic patient's body after administration) than rapid-acting insulin. Therefore, long-acting insulin is an example of a type of insulin that has a longer duration of activity than types of insulin such as rapid-acting insulin.

[0092] Figure 1 shows an example of process 100 for guiding PWD in managing their own insulin therapy. Process 100 can be used in combination with one or more other examples described elsewhere in this specification. Process 100 is schematically shown as multiple boxes. Some implementations may perform more or fewer actions than shown.

[0093] Pre-training survey stage 102 involves collecting data from and / or about the PWD. Data can be collected by conducting one or more surveys with the PWD. For example, the survey can be conducted in person and / or presented on a display (e.g., on a mobile electronic device) or using a computer system speaker. The survey may inquire about the PWD's goals regarding the management of insulin therapy and other aspects of the PWD's person or style. For example, the survey may examine how to communicate with the PWD during insulin therapy. Pre-training survey stage 102 can also collect information from the PWD's healthcare provider (HCP). For example, the HCP may be a physician, nurse, other health professional, or a family member or other caregiver. Information collected from the HCP may include information about the PWD's health status or how the PWD should be treated. For example, the HCP may recommend raising the PWD's health status to an optimal blood glucose level or modifying one or more aspects of process 100. As another example, pre-training assessment stage 102 may attempt to determine whether to apply a general initiation regimen for insulin therapy to PWD in terms of therapeutic adequacy or messaging receptivity.

[0094] Process 100 may include a user training stage 104 which may be directed to the PWD and / or the PWD's HCP. In some implementations, the user training stage 104 provides education on diabetes and / or insulin therapy, as well as on other aspects of Process 100. The user training stage 104 may be conducted in person and / or presented on a display (e.g., on a mobile electronic device) or using a computer system speaker. The format and / or content of the user training stage 104 may be adjusted using data or other information collected in the pre-training survey stage 102.

[0095] Process 100 may include a high-touch interaction phase 106. For example, "high-touch" could mean that the high-touch interaction phase 106 involves relatively intensive, frequent, or substantial engagement with the PWD. In some implementations, the high-touch interaction phase 106 can be characterized as an initial phase in that it precedes subsequent phases of less intensive / less frequent / substantial interaction with the PWD (as described below). Information collected in the pre-training survey phase 102 can modify the content of the high-touch interaction phase 106 or how the high-touch interaction phase 106 is conducted. For example, based on the collected information, process 100 may identify other PWDs with similar medical and / or physiological characteristics to the current PWD and use their settings or other process configurations as a starting point for the PWD. As another example, if no other PWDs with similar characteristics are available, process 100 may use data obtained from or for the PWD as the initial settings.

[0096] In some implementations, some or all aspects of the high-touch interaction phase 106 can be characterized as a titration of the PWD and the PWD's insulin therapy. The titration may involve the use of multiple insights (exemplified below), and the intended action performed is the adjustment of insulin therapy. Initially (e.g., within the high-touch interaction phase 106), the titration may involve a set of high-touch interactions about the PWD within a relatively short period (e.g., days or weeks) after the start of the high-touch interaction phase 106, designed to establish appropriate data and build conditions for rapid adjustment of insulin therapy. While the insight engine (exemplified below) may be designed to operate primarily with available data, the high-touch interaction phase 106 may be an exception. Rather, in the high-touch interaction phase 106, the PWD may be asked to record meals, wake-up times, or whether they missed any doses. Process 100 may, in addition or instead, use other sources of information to enhance these observations. For example, a wearable device may provide information for data collection.

[0097] The dialogue during the high-touch dialogue phase 106 may include the establishment of one or more insulin dose settings through iterative adjustments of insulin therapy. For example, these may be considered initial settings for the purpose of initiating an optimization procedure in which the settings may be changed one or more times. This may be characterized as part of the titration of insulin dosage for PWB. The insulin settings may include, but are not limited to, basal dose settings 108, meal dose settings 110, and corrective dose settings 112.

[0098] In some implementations, basal insulin dose settings for PWDs can be established by American Diabetes Association (ADA) guidance, which is widely accepted and utilized as an example of publicly available guidelines on insulin therapy. Another example of guidelines is the recommendations from the American Association of Clinical Endocrinology (AACE). These guidelines can be applied by directly utilizing blood glucose values ​​from a blood glucose meter to understand fasting plasma glucose (FPG) levels in PWDs. The guidelines can be applied by interpreting CGM values ​​to reduce the user's burden during this critical period when starting a new therapy. In some implementations, CGM can be used to assess minimum morning glucose (MMG) in PWDs. For example, MMG can be defined as the lowest blood glucose level between 4 a.m. and 30 minutes before the first activity of the day. These values ​​can be processed in one or more ways. For example, elevated levels due to treatment of nocturnal lows (e.g., via eating) can be excluded. As another example, mornings with rebound elevated levels resulting from nocturnal lows can be excluded. As yet another example, adjustments can be made for non-standard wake times (e.g., for shift workers). A basic, consistent, and flexible framework for initiation can be constructed, independent of a given glucose measurement tool. Built from guidelines, PWD can be initiated daily at bedtime with a predetermined number of units, or a specific amount indicated by HCP. The system can recommend increasing by a specific number of units or a certain percentage at regular intervals (e.g., once every certain number of days) until FPG and / or MMG are within the target range. Other adjustments can be made. For example, the system can assess basal excess and consider adjunctive therapy (e.g., whether the basal dose exceeds a predetermined number of units / kg / day, whether the bedtime-morning (BeAM) glucose difference and / or the post-prandial-pre-prandial difference is elevated, whether hypoglycemia (recognized or unrecognized) and / or high variability (e.g., the coefficient of variation exceeds a pre-specified number).

[0099] In some implementations, the mealtime insulin dose setting 110 for PWD can be established using a stepwise approach that focuses on one meal at a time. The mealtime dose setting 110 can take into account the mean measurements of blood glucose (e.g., A1C test), basal dose setting 108, FPG, and / or MMG. Pre-meal blood glucose levels can be taken into account. The system can recommend increasing by a specific number of units or a certain percentage at regular intervals (e.g., once every certain number of days) until the pre-meal blood glucose level for the next meal falls within the target range. If hypoglycemia occurs, the system can examine the blood glucose pattern and reduce the insulin dose by a predefined amount.

[0100] In some implementations, the insulin correction dose setting for PWD112 can be established to compensate for mean blood glucose levels (e.g., A1C) that are outside the target range. The system can take insulin sensitivity factors for PWD into consideration.

[0101] As described above, some or all settings in the high-touch dialogue phase 106 (e.g., basal dose setting 108, meal dose setting 110, and / or corrected dose setting 112) can be established based on publicly available guidelines for insulin therapy. This can be done after the PWD's HCP has confirmed that there is no need for individualized glycemic targets for the PWD. Such guidelines may specify one or more target ranges for A1C (e.g., considering that the PWD is not pregnant and does not have significant hypoglycemia), the time within the range target can be assessed using portable glucose profiles and / or glucose management indicators, and the system may accept A1C levels lower than the target based on HCP judgment and PWD preferences while monitoring for significant hypoglycemia or other adverse effects, and the system may apply less stringent targets (e.g., for A1C) if the harm of treatment outweighs the benefits. The system may set glycemic targets based on individual criteria, but a comprehensive clinical target can be defined as the default. For example, this may include peaks in A1C levels, pre-meal capillary plasma glucose levels, and / or post-meal capillary plasma glucose levels. A somewhat stringent blood glucose target may be appropriate for individual patients. Continuous glucose monitoring (CGM) can be used to assess the blood glucose target.

[0102] The duration of the high-touch interaction phase 106 can be defined in one of several ways. In some implementations, a goal-based duration (e.g., an outcome-based duration) can be used. The system can determine that it has had sufficient interaction with the PWD and has sufficient data about the PWD so that the dose is the correct dose. For example, the dose can be brought close enough to the correct level so that a non-interactive phase, which relies on reinforcement learning, can be initiated. The system's goal-based assessment may take into account that blood glucose remains within the target range for at least a certain percentage of time, and that the PWD is taking insulin as recommended. In some implementations, a predetermined duration (e.g., a certain number of days) can be used.

[0103] Process 100 may include a progression phase 114. Progression phase 114 includes collecting performance metrics for the PWD, applying reinforcement learning to the performance metrics, and automatically performing actions on the PWD based on the reinforcement learning. Progression phase 114 can be performed based on survey information from the pre-training survey phase 102. For example, specific insulin therapy recommendations and / or methods of communication with the PWD may influence whether to continue monitoring diabetes and watch for new developments, to intensively manage diabetes, to minimize costs and expenses for the PWD, to maximize health benefits for the PWD, and how to talk to the PWD. In progression phase 114, the system can look for the most influential behaviors and insulin therapy settings, as well as what to communicate to the PWD. For example, the message might be that the PWD is not taking sufficient doses, rather than recommending each dose that should be taken. As another example, the PWD might be recommended to take doses earlier or later. In some implementations, one or more insights may be evaluated, potentially leading to the execution of at least one action. Actions may include communication with the PWD (or HCP) and / or changes to insulin therapy settings. The message generated for the PWD can be selected considering the PWD's preferences. After taking action based on an insight, that insight can be monitored for a period of time. During that period, if the condition improves, the PWD can be notified and monitoring can be terminated. If no improvement is observed, other remediation measures can be considered.

[0104] Progression Phase 114 may include one or more ongoing therapy adjustment periods. These ongoing therapy adjustment periods may be intended to adapt to the needs of the patient throughout the remainder of the PWD's treatment course using multiple daily injections (MDIs) of insulin. The ongoing therapy adjustment periods may be built around the investigation of performance metrics constructed to understand the PWD's behavioral, physiological, and drug administration behaviors. These metrics can be kept as up-to-date as possible, updated between once daily and once hour. For example, performance metrics are not intended to be real-time, and many observations look for recent historical trends rather than sudden or real-time changes. (Examples below) An insight generation engine can sequentially evaluate insights and, if any, determine which insights to act upon. The most important negative insights can be considered, and only the most relevant (i.e., most likely to yield positive outcomes) insights can be presented to the PWD. The most positive compliments can be considered. An insight may be triggered, and action may be taken (e.g., a coaching message may be sent to the PWD, or therapy may be adjusted), after which an observation period may begin, and a cool-down period may start. The observation period is a period during which the system can look for improvement. The criteria for knowing that an insight has improved may be, but not necessarily, the same as one or more criteria that triggered the insight in the first place. The system may use other criteria or specially constructed classification methods to determine whether the PWD has improved based on the actions taken. A cool-down period may be used to ensure that the system does not act on insights that have recently worked. Safety-related insights (e.g., those involving hypoglycemia) may have the highest priority, and cool-down periods may range from very low to none.Insight may not address safety issues, but rather coach non-urgent issues that tend to lead to insulin reduction, such as excessive hypoglycemia observed despite normal or low insulin delivery.

[0105] Both titration and coaching can be included in this framework. For example, titration primarily involves actions that change the setting of insulin therapy. As another example, coaching primarily involves actions that instruct the PWD on how to improve their behavior (or where to obtain sensors or insulin, if that is the issue).

[0106] Action / coaching procedures can push questions to the PWD that help the system provide better coaching or instructions. For example, communication could state, "I realized I'm not wearing my sensor. Do you need help with (a) Rx, (b) arranging for a sensor, (c) dealing with a sensor malfunction / sensor detachment, (d) allergic reaction to sensor adhesive, (h) taking a break from the sensor?" The response can help build and shape the coaching. "I realized I'm not taking my <long-acting insulin brand>. How can I help you with (a) Rx, (b) getting affordable insulin?" The PWD can input into the system (e.g., by using a device to text, selecting from options, or speaking into a microphone).

[0107] Once an action is taken, an observation period can be implemented. The observation period facilitates evaluation of how effective the action taken was in improving the PWD treatment process (the PWD user's state). By evaluating how effective messaging / actions were, the system can change the likelihood of taking a given action in the future. For example, changing messaging preferences (e.g., PWD responding to cost-related messages) will make future messages more likely to focus on areas that worked well in the past. Similarly, threshold, cooldown, and observation period parameters may be adaptive depending on successful actions.

[0108] Parameter adjustments can be adapted, in addition to or instead, based on similar patients. In some implementations, a given user may not require many insulin titration events after the initial titration period. However, if similar patients are seeing success with titration actions, the system can adjust the parameters for all similar patients to capitalize on this learning. Other aspects besides titration can also be learned, or instead, from data of other patients. This may include, but is not limited to, coaching messages and / or the balance between basal and bolus doses.

[0109] Progression phase 114 may include one or more adjustments to one or more of the basal dose setting 108, meal dose setting 110, and / or corrected dose setting 112. In some implementations, process 100 may recommend increasing, decreasing, or keeping the basal dose setting 108 unchanged during progression phase 114, taking into account one or more of the following: the average glucose value in a given time interval, the percentage of time spent hypoglycemia and hyperglycemia in a given time interval, the calculated glucose change rate in a given time interval, the treatment of hypoglycemia that occurred in a given time interval, or hypoglycemia that occurs beyond a given period of time. Reinforcement learning may define positive and / or negative rewards for one or more situations. A target state may be considered when recommending increasing, decreasing, or keeping the basal dose setting 108 unchanged. For example, a target state may be defined based on the average nighttime blood glucose level and the absence of hypoglycemia or hyperglycemia. If more than a certain percentage of days exhibit hypoglycemia, the system may find a recommendation by considering only the hypoglycemic days. If hypoglycemia is found to be triggered by dinner or a late-night snack bolus, the system may ignore hypoglycemia for that day in its assessment. In some implementations, a positive reward can be generated for the system when PWD reaches the target state and / or when the change in hypoglycemia, hyperglycemia, or mean glucose error decreases or disappears.

[0110] In some implementations, process 100 may recommend increasing, decreasing, or keeping unchanged the meal dose setting 110 during progression phase 114, taking into account one or more of the following: minimum blood glucose level during a specified time interval, target blood glucose level, glucose change rate calculated during a specified time interval, hypoglycemia occurring beyond a specified time, or no hypoglycemia occurring. Reinforcement learning can define positive and / or negative rewards for one or more situations. A target state can be defined. For example, the target state may include postprandial error within a specified range and no hypoglycemia. In some implementations, a positive reward can be generated for the system if PWD reaches the target state, and / or if the change in the rate of change and postprandial error is decreased or not at all. In some implementations, a negative reward can be generated for the system if hypoglycemia is between specified levels. If more than a certain percentage of days exhibit hypoglycemia, the system may find a recommendation considering only the hypoglycemic days.

[0111] In some implementations, process 100 may recommend increasing, decreasing, or keeping the corrected dose setting 112 unchanged during progression phase 114, taking into account one or more of the following: the minimum blood glucose level during a specified time interval, the glucose change rate calculated during a specified time interval, or the blood glucose risk index. Reinforcement learning can define positive and / or negative rewards for one or more situations. A target state can be defined. For example, the target state may include corrected error within a specified range and no hypoglycemia. In some implementations, a positive reward can be generated for the system if PWD reaches the target state, and / or if the change in the rate of change and corrected error is decreased or not at all. In some implementations, a negative reward can be generated for the system if hypoglycemia is between specified levels. If more than a certain percentage of days exhibit hypoglycemia, the system may find a recommendation considering only the hypoglycemic days.

[0112] The system can use various statistical tools in the evaluation during progression phase 114. For example, the mean and / or median can be used for one or more measured entities.

[0113] When evaluating whether to adjust a comprehensive therapy, one or more considerations may be taken into account. Considerations regarding escalation, or unlikely regressions between therapies, can be defined. For example, this could include escalation from basal-only to MDI. Another example could involve escalation from the use of a specific drug (e.g., glucagon-like peptide-1) to the use of a drug with a basal dose. Considerations regarding sharing parameters (e.g., insulin sensitivity factors) between algorithms can be defined.

[0114] Figure 2 shows an example of the pre-training survey stage 102 of process 100 in Figure 1. The pre-training survey stage 102 may include a user survey 200. For example, the user may fill out the user questionnaire 200 by typing and / or speaking. In some implementations, the PWD is asked several questions to determine their thoughts about their current therapy. The answers to these questions can be used to determine the PWD's psychological profile. The system can determine how much the PWD fears hypoglycemia and what is driving the PWD's decisions about their therapy (e.g., cost, time, support from family, understanding of how diabetes is treated, etc.). This can facilitate the making of one or more settings 202. For example, settings 202 may include setting therapy targets, a mental model of diet, defining contact frequency, defining messaging focus, and / or specifying assertiveness regarding communication / actions.

[0115] The pre-training survey stage 102 may include an HCP survey 204. The HCP survey 204 may request the HCP to set therapeutic parameters. This can facilitate the making of one or more settings 206. For example, settings 206 may include setting glucose targets, long-acting (LA) insulin types, rapid-acting (RA) insulin types, RA initial therapy settings, high-touch phase orders, and / or standing orders.

[0116] The pre-training survey stage 102 may include defining a therapy regimen 208. The therapy regimen 208 may include a set of parameters defining the therapy. These may include insulin dose and mental model, as well as / or the assertiveness of therapy adjustments, target glucose, contact frequency, communication method, and coaching focus. This may facilitate the construction of one or more state action tables 210. For example, a state action table 210 may define how performance metrics lead to actions / recommendations.

[0117] Figure 3 shows an example of a patient messaging profile 300 that can be constructed in process 100 of Figure 1. The patient messaging profile 300 can be used in combination with one or more other examples described elsewhere in this specification. The patient messaging profile 300 includes a scale 302 that can define the profile content. The scale 302 can specify probabilities. For example, the probabilities may reflect the likelihood that a particular type of message will improve an observed condition. The patient messaging profile 300 can include one or more entries 304 measured against the scale 302. For example, an entry 304 may specify one or more of the following: saving money, long-term health, improved mood, risk of hypoglycemia, and / or the sense that the diabetes care provider (DCP) understands the patient's condition (PWD). Other approaches can be used.

[0118] Based on the assessments reflected in the Patient Messaging Profile 300, recommended goals will be provided to the PWD. Because the PWD's situation may change over time, the PWD may be able to modify the recommended goals (e.g., within the app). While the PWD's goals may not change the type of clinical insight or the approach to the clinical insight, they may change the messaging regarding how the clinical insight is expressed. A variety of different goal descriptions can be used. Goals can be individualized based on the duration of diabetes, age / expected lifespan, comorbidities, known cardiovascular disease or advanced microvascular complications, asymptomatic hypoglycemia, and individual patient considerations. Examples of goals include, but are not limited to, improving my overall long-term health, reducing hypoglycemia and managing associated stress levels, feeling more relaxed, becoming more active and focused, learning more about my diabetes and receiving more personalized therapy, or reducing overall healthcare costs. The system then utilizes standard guidelines (e.g., from the ADA or AACE) for titration during the initiation of therapy, starting with basal therapy, based on glycemic targets.

[0119] Figure 4 shows an example of a system 400 for diabetes management. System 400 can be used in combination with or in embodiments thereof one or more other examples described elsewhere in this specification. Here, system 400 includes an insulin injection pen 402. In some implementations, insulin injection pen 402 is considered to be an insulin injection pen for a specific type of insulin, e.g., a rapid-acting insulin pen. For example, insulin injection pen 402 is a Humalog® pen, a Novolog® pen, or an Apidra® pen. Here, system 400 includes an insulin injection pen 404. In some implementations, insulin injection pen 404 is considered to be an insulin injection pen for a specific other type of insulin having a longer activity time than the activity time of a specific type of insulin, e.g., a long-acting insulin pen. For example, insulin injection pen 404 is a Lantus® pen, a Levemir® pen, a Toujeo® pen, or a Tresiba® pen. Here, system 400 includes a glucose monitor 406 (e.g., a CGM) or another glucose sensor, and a remote user interface device 408. As illustrated, each insulin injection pen 402 and 404 includes corresponding pen caps 410 and 412, respectively, which communicate wirelessly with other components of system 400. As illustrated, insulin injection pens 402 and 404 may each include dials 414 and 416 for the user to set the dose to be delivered, and corresponding dose indicator windows 418 and 420. In some implementations, one or more of the insulin injection pens 402 and / or 404 may include dose capture technology and / or communicate wirelessly with other components of system 400. Additional details regarding expected insulin pens and / or insulin pen caps are illustrated below.

[0120] The glucose monitor 406 or another glucose sensor may include any suitable sensor device and / or monitoring system capable of providing data that can be used to estimate one or more blood glucose levels. As illustrated, the glucose monitor 406 or another glucose sensor may be a sensor configured to wirelessly transmit glucose data. For example, the glucose monitor 406 or another glucose sensor may include optical communication devices, infrared communication devices, wireless communication devices (such as antennas), and / or chipsets (such as Bluetooth devices (e.g., Bluetooth Low Energy (BLE), Classic Bluetooth, etc.), NFC devices, 802.6 devices (e.g., Metropolitan Area Network (MAN), ZigBee devices, etc.), WiFi devices, WiMax devices, Long-Term Evolution (LTE) devices, cellular communication equipment, etc.), and / or similar. The glucose monitor 406 or another glucose sensor may exchange data with a network and / or any other devices or systems described herein. In some cases, the glucose monitor 406 or another glucose sensor can be queried via an NFC device by the user moving one or more components of the system 400 close to the glucose monitor 406 or another glucose sensor to power the glucose monitor 406 or another glucose sensor and / or to transmit glucose data from the glucose monitor 406 or another glucose sensor to other components of the system 400. For example, the pen caps 410 and / or 412 can exchange data with the glucose monitor 406 or another glucose sensor (e.g., obtain glucose values ​​from it) by bringing them close to the glucose monitor 406 or another glucose sensor.

[0121] As illustrated, in some examples, the remote user interface device 408 is a mobile electronic device (e.g., a smartphone). In some implementations, any suitable remote user interface device can be used, including but not limited to computer tablets, smartphones, wearable computing devices, smartwatches, fitness trackers, laptop computers, desktop computers, smart insulin pens (e.g., pen caps 410 and / or 412), and / or other suitable computing devices. As shown in the exemplary user interface of an exemplary mobile app running on an illustrated smartphone, the user interface may include a bolus calculator button 422 and, optionally, other buttons for the user to enter data or request recommendations. The exemplary user interface may also include, or instead, a display of (e.g., historical, current, and / or predicted) blood glucose data. As illustrated, the user interface includes a graph 424 of historical data (e.g., from the last 30 minutes), a continuation of that graph 426 with predicted data, a point indicator 428 showing the current (or latest) estimated blood glucose level, and a display 430 of the current (or latest) estimated blood glucose level. The user interface may, in addition to or instead, include text 432, text 434 describing glucose data, and / or text 436 providing suggested actions. For example, text 434 may provide suggestions for insulin, carbohydrates, or other therapies. For example, text 436 may suggest that the user retrieve glucose data. In some cases, the user interface may allow the user to click on glucose data or navigate within the mobile app to retrieve more detailed or complete blood glucose data. In some implementation forms, the user interface may, in addition to or instead, provide one or more insights for the benefit of diabetic patients.For example, Insight can deliver a message along the lines of, "You usually have a meal bolus between noon and 2 p.m., but you didn't have one today, and your blood sugar is elevated. Did you forget to have a bolus?"

[0122] The user interface can display insulin data. In some cases, the user interface can display the amount of IOB438 which may only be applicable to a specific type of insulin (including, but not limited to, rapid-acting insulin). In some implementations, IOB (sometimes called active insulin) can be defined as the amount of insulin delivered that remains active in the human body based on the estimated duration of insulin action. In some cases, IOB calculations can be performed for both rapid-acting and long-acting insulin. In some cases, the user interface can display information440 including, but not limited to, the time and / or amount of the latest dose of rapid-acting insulin and / or long-acting insulin. In some cases, the user interface can allow the user to click on insulin data or navigate within the mobile app to obtain more detailed and / or more complete insulin delivery data. In some cases, the user interface can overlay blood glucose data and insulin delivery data in any preferred format, such as a graphical display of the timing of blood glucose data versus the timing of insulin delivery data.

[0123] When in use, the user (e.g., PWD and / or HCP) can use system 400 to obtain recommendations regarding an appropriate insulin dose. If it is necessary to deliver long-acting insulin in the future, text 434 can be modified to provide a recommended long-acting insulin dose. In some cases, the recommended dose may appear on pen caps 410 and / or 412. If the user wishes to deliver a bolus of rapid-acting insulin, the user can press bolus calculator button 422 to enter the bolus calculator. Any suitable bolus calculator can be used in the systems, methods, and devices provided herein. For example, the bolus calculator may provide a user interface for the user to input a meal announcement as either correction only, small meal, normal-sized meal, or large meal. Once a meal size is selected, the user interface may provide a recommended bolus dose based on the number of carbohydrates associated with the corresponding button, and optionally based on blood glucose data. In addition, or instead, the dose capture pen caps for insulin injection pens 402 and / or 404 may include a user interface that allows the user to obtain meal bolus recommendations for different types of meals (small, medium, large meals; breakfast, lunch, dinner; 10 grams of carbohydrates, 20 grams of carbohydrates, 45 grams of carbohydrates, etc.) and / or be announced meal sizes, including but not limited to small, medium, or large meals. In some implementations, the texts 432, 434, and / or 436 are presented elsewhere than on the user interface device 408. For example, the presentation may be on the pen caps 410 and / or 412.

[0124] The pen caps 410 and / or 412 may include at least one display. In some implementations, the pen cap 410 includes a display 442. In some implementations, the pen cap 412 includes a display 444. The displays 442 and / or 444 may include any preferred type of display technology, including but not limited to dynamic electronic displays (e.g., light-emitting diode displays) or static electronic displays (e.g., E-ink displays). The displays 442 and / or 444 may present information to the user, including but not limited to any information outputs described elsewhere in this specification. For example, they may present insulin dosage suggestions and / or alerts or other communications.

[0125] The pen caps 410 and / or 412 may include at least one input control. In some implementations, the pen cap 410 includes a button 446. In some implementations, the pen cap 412 includes a button 448. The buttons 446 and / or 448 may include any preferred type of input technology, including but not limited to electronic switches. The buttons 446 and / or 448 may trigger the corresponding pen caps 410 and / or 412 to perform one or more actions, including but not limited to any actions described elsewhere in this specification.

[0126] The pen caps 410 and / or 412 can record and / or transmit one or more types of pen capping information. Pen capping information (e.g., information about when the pen cap was attached to and / or released from the injection pen) may include information about the current capping period (e.g., time since the last capping), information about the duration of one or more uncappings (which may also be referred to herein as “decappings”), and / or the timing of each uncapping and each capping (e.g., time or elapsed time). For example, the capping information may include data reflecting when the pen cap was placed on the insulin injection pen, data reflecting when the pen cap was removed from the insulin injection pen, or both. In some embodiments, the pen capping information may be presented to the user on a display on the pen cap. In some embodiments, the pen capping information may be presented by a speaker inside the pen cap. For example, in some embodiments, the pen cap may provide a timer clock that counts up from the time the pen cap was last attached to the injection pen. In some embodiments, the pen cap accessory can wirelessly transmit pen capping information to a remote computing device (e.g., a user interface device 408 and / or a smartphone, tablet, etc.). In some embodiments, one or more accessories or smart delivery devices can detect other events associated with the drug delivery action and use that information in the manner described herein with respect to pen capping information. For example, in some cases, an injection pen accessory may be attached to an injection pen so that it can detect the mechanical movement of the administration mechanism and determine the drug administration time (and, but not necessarily, optionally, the amount of drug delivered at that time).

[0127] Pen-capping information can be stored, displayed, and / or analyzed in combination with glucose data to determine user behavior, such as whether the person is administering insulin appropriately for meals, and / or to correct elevated blood glucose levels. In some embodiments, pen-capping information may be presented on a graphical representation of glucose data for the user and / or to the user and / or healthcare professionals. In some embodiments, glucose data from the period following each capping event may be evaluated to determine whether the user administered insulin appropriately for that capping event, e.g., an appropriate dose, an underdose, or an overdose.

[0128] In some embodiments, a pen-capping event may be ignored if other information indicates that no dose was provided. For example, the event may be ignored if no change in the dosage selection of the insulin pen (e.g., dial) was detected. In some embodiments, pen uncapping and recapping events may be ignored if the total uncapping time is less than a first threshold (e.g., 4–6 seconds). For example, the threshold may be determined by setting it to a time amount that is too short to allow an injection, but long enough to allow the user to check the end of the pen to see if there is any insulin remaining or if there is a needle attached to the pen. In some cases, the total decapping time for a decapping event (the time between the uncapping event and the subsequent recapping) may be analyzed in combination with blood glucose data to determine whether an injection occurred during that decapping event. In some cases, if the total decapping time exceeds a second threshold period (e.g., at least 15 minutes, at least 30 minutes, etc.), the approximate time of the injection may be determined using glucose data and the methods described herein for detecting meal timing.

[0129] Some or all components of system 400 can perform one or more operations or functions as described elsewhere in this specification. In some implementations, system 400 can evaluate insights based on reinforcement learning and perform one or more actions for triggered insights. System 400 can determine insulin use by detecting the removal and / or replacement of pen caps 410 and / or 412. System 400 can detect the amount of insulin injected.

[0130] Figure 5 shows an example of a system 500 for guiding diabetic patients in managing insulin therapy. System 500 can be used in combination with, or include in its embodiments, one or more other examples described elsewhere in this specification. System 500 is shown as comprising several distinct components. The functionality of system 500 can be implemented using more or fewer components. System 500 includes an insight generation engine 502, which can be implemented using components described below with reference to Figure 12. The insight generation engine 502 evaluates several insights based on reinforcement learning and, for each insight, decides whether to generate that insight. Each insight can be associated with performing one or more actions. For example, performing an action may include providing a message to PWD and / or automatically changing insulin therapy. This insight can be designed to cover the entire treatment process of PWD with diabetes. The insight may be accompanied by observations from data, which should be invoked so that an action can be taken. The insight should be sufficiently relevant so that, if the appropriate action is taken, the PWD's MDI therapy can be significantly improved.

[0131] An insight can describe a suboptimal MDI therapy state, followed by an action. The insight generation engine 502 can evaluate the insight conditions and determine which insights meet the trigger criteria. Triggered insights are followed up with actions. To evaluate whether the actions were successful, system 500 can monitor PWD during the observation period. If the condition improves, system 500 can give the insight generation engine 502 a positive reward and update parameters so that the insight generation engine 502 is more likely to take such actions in the future. If the condition does not improve, system 500 can give the insight generation engine 502 a negative reward and update parameters so that the insight generation engine 502 is less likely to take such actions in the future. After the observation period, system 500 may not allow insights to be triggered for a while, a scenario that can be called a cool-down period. This ensures that system 500 does not overwhelm PWD with too many actions. If the PWD status does not trigger an Insight for an extended period, the Insight generation engine 502 may trigger a streak reward to commend its achievement.

[0132] The insight may have multiple parameters in the memory unit 504, which is accessible from the insight generation engine 502. These parameters may include, but are not limited to: trigger conditions, such as a set of performance metric-based conditions indicating that coaching action should be taken if, for example, less than a certain amount of RA dose is taken per period and the time PWD spends within the blood glucose range is less than a threshold time; coaching actions, such as a set of possible actions that can be taken to help improve PWD, such as a message or therapeutic adjustment, for example, to present a card with PWD's postprandial glucose when PWD takes an RA dose; and observation windows for measuring how effective an action is, such as specifying how long to observe before determining whether the intervention was successful (some interventions may take time to judge from performance metric data, and therefore this...). The parameters may allow System 500 to look at the situation over a longer period before determining whether the intervention worked; a cool-down period to avoid repeatedly telling the PWD that it is not behaving properly, the cool-down period may be defined to ensure that insights are not frequently triggered (with regard to therapeutic adjustments, System 500 may also limit the rate of adjustment, such as setting a two-week cool-down period, to ensure that System 500 is not too quick to consider another therapeutic adjustment), and more than one cool-down period may be associated (for example, triggering a particular insight may initiate a cool-down for multiple insights); a reward condition, for example, a condition under which System 500 rewards the absence of triggered insights over a period of time, for example, by presenting a message such as, "Congratulations! You have taken <long-acting brand name> at the scheduled time every day for the past 60 days!"The memory unit 504 can reflect the PWD status regarding the insight. Since insight generation is based on knowing the observation period and cooldown period, the user state is considered before the insight is determined.

[0133] System 500 can utilize data about PWD or data related to PWD. The data can be provided in one of several ways, for example, by a data lake 506. In some implementations, the data lake 506 may be a centralized repository for storing, processing, and protecting large amounts of structured, semi-structured, and / or unstructured data. For example, the data lake 506 may be provided by a cloud infrastructure that collects sensor readings, smart cap removal / replacement, user input, and / or other information about PWD. Based on the information from the data lake 506, at least one calculation 508 can be performed within System 500. In some implementations, performance metric calculations can be performed. Performance metrics can be calculated periodically, such as daily. Examples of performance metrics include, but are not limited to, the number of times the smart cap of a first injection pen is removed or replaced per day (e.g., using long-acting insulin), the number of times the smart cap of a second injection pen is removed or replaced per day (e.g., using rapid-acting insulin), the time within the specified range, or the postprandial glucose level. Performance metrics can be stored in a performance metric database 510 accessible by the insight generation engine 502.

[0134] The system 500 may include a component 512 that presents the output from the insight generation engine 502. In some implementations, the output of component 512 is specific to the PWD. For example, the output may include PWD-specific coaching and / or titration output.

[0135] Figure 6 shows an example of a state diagram 600 that can be used by the insight generation engine 502 of Figure 5. The state diagram 600 can be used in addition to, or instead of, one or more other examples described elsewhere in this specification. The state diagram 600 represents different states for at least one insight, each of which is schematically represented as a circle and arrows indicate possible state transitions. The state diagram 600 includes a non-triggered state 602. For example, this could be the default initial state for each insight. The triggered state 604 indicates that a decision has been made to trigger an insight. For example, an insight is triggered based on an evaluation of a performance metric that indicates that the criteria for the insight are met. The action state 606 indicates that at least one action associated with the triggered insight is performed in response to the trigger. The observation state 608 indicates that the triggered insight is under observation after being triggered. The observation period can be used to collect information used for reinforcement learning. The positive feedback reward state 610 indicates that the system is given positive feedback based on observations related to the triggered insight. The negative feedback reward state 612 indicates that the system is given negative feedback based on observations related to the triggered insight. During the observation state 608, the positive feedback reward state 610, and the negative feedback reward state 612, the triggered insight may not be eligible to be triggered. The observation state 608 includes the time over which the effect of the action was observed after the insight was triggered. For some insights, the system may look for performance metrics to change to an acceptable level. For other insights, the system may use specialized machine learning methods to determine the success of the action. For example, a machine learning method that examines glucose patterns may be used to determine whether to refresh long-acting insulin to confirm that the insulin dosage has changed.

[0136] The cool-down state 614 indicates that the triggered insight is in a cool-down period after the observation state 608. During the cool-down state 614, the triggered insight may not be eligible to be triggered. After the cool-down state 614, the insight may again enter the non-triggered state 602. During the non-triggered state 602, for each insight, an assessment can be made as to whether the insight has not been triggered for at least a threshold time period. Such a period can be considered a favorable or desirable situation for the PWD in that the PWD's diabetes was under control during that period and no change in this particular coaching or therapy was required. The streak state 616 indicates that the insight has not been triggered for at least a threshold time period and has continued to be successful. For example, the streak state 616 could include presenting a commendation message to the PWD.

[0137] Figure 7 schematically shows the insight prioritization 700 that can be performed by the insight generation engine 502 of Figure 5. Prioritization 700 can be used in addition to, or instead of, one or more other examples described elsewhere in this specification. Prioritization 700 is performed on insights 702-1 to 702-N, where N = 2, 3, ... Here, PWD is referred to as user X. For example, insights that may affect PWD's lifespan may be given the highest priority. As another example, insights that may affect PWD's health may be given a lower priority (e.g., second priority) than insights that affect life, but a higher priority than at least one other insight. As yet another example, insights that do not affect PWD's life or health, but relate to improving or optimizing PWD's diabetes care, may be given a lower priority than insights that affect health. Based on prioritization 700, insight 702-1 can be considered the highest priority insight. Insight 702-2 can be considered the second highest priority insight. Insight 702-N can be considered the lowest priority insight. Therefore, multiple insights can be ranked in prioritization 700 and evaluated in that order.

[0138] Insight competition can occur. Insight competition can be a process in which the system (e.g., the insight generation engine 502 in Figure 5) determines the most relevant insights. That is, insights may have a natural priority compared to one another (e.g., as described above). Nevertheless, given the current state of PWD and data, multiple positive actions (e.g., praise and streaks) and negative actions (coaching or therapeutic adjustments may be needed) may be possible. Insight competition can select the right insights to present to PWD. This ensures a balance in the number and types of insights presented at one time.

[0139] Figures 8A and 8B show examples of processes 800 that can be performed by the insight generation engine of Figure 5. Process 800 can also be used in one or more other examples described elsewhere in this specification. More or fewer operations can be performed than shown. Unless otherwise indicated, two or more operations can be performed in different orders. Process 800 can be run repeatedly to evaluate each of several insights in their prioritization order. Here, process 800 labels the insight currently being evaluated as the highest priority insight, which has not yet been evaluated in previous iterations of process 800.

[0140] In operation 802, a determination is made as to whether the performance metrics (for example, in the performance metrics database 510 in Figure 5) provide sufficient data. The criteria for sufficient data may vary depending on the insight. For example, if the system has not seen a sufficient number of lunchtime events for the PWD over a past period to reliably assess what happens to the PWD at lunchtime, the lunchtime insight may be skipped. If the result of operation 802 is no, the process in operation 804 may proceed to the next insight in order of priority. For at least one insight, a lack or absence of performance metrics may satisfy the conditions for triggering the insight. For example, if there is too little data in the system, a coaching message may be generated to encourage the PWD to use the system more so that data related to diabetes management is obtained.

[0141] If the result of operation 802 is yes, operation 806 can determine whether the current insight is being observed (for example, observation state 608 in Figure 6). If the insight is not being observed, it becomes eligible for trigger evaluation (however, it may not be triggered depending on the circumstances). Accordingly, if the result of operation 806 is no, operation 810 can determine whether the performance metric will trigger the insight. The determination may take into account one or more settings 812 related to the insight, such as thresholds.

[0142] If an insight is not triggered by the current state of the PWD, the PWD may continue to be in a state that does not trigger this particular insight. If the result of action 810 is no, action 814 can determine whether the insight has not been triggered over a specified period of time. The determination may take into account settings 816, such as one or more thresholds for streak praise. If the result of action 814 is no, in action 818, process 800 may proceed to the next insight in order of priority. If the result of action 814 is yes, in action 820, a streak praise may be performed. A streak praise may be performed based on settings 822, such as one or more styles or methods for coaching.

[0143] If the result of action 810 is yes, action 824 can determine whether the insight is currently in a cool-down period (e.g., cool-down state 614 in Figure 6). The determination may take into account one or more settings 826, such as thresholds, related to the insight. If the result of action 824 is yes, in action 828, process 800 can proceed to the next insight in order of priority. If the result of action 824 is no, in action 830, process 800 can perform one or more actions associated with the triggered insight. For example, the action may involve adjustments to coaching and / or therapy. The action may take into account one or more settings 832, such as one or more styles or methods for coaching. After performing the action, in action 834, observation of the insight can begin (and process 800 can proceed to evaluate the next insight in order of priority).

[0144] Returning to action 806, if the result of this determination is "yes," action 836 may determine whether one or more improvement criteria have been met. This determination may take into account settings 838, such as one or more improvement thresholds for the insight. If the result of action 836 is "no," action 840 may determine whether the end of the observation window has been reached. This determination may take into account settings 842, such as the definition of the observation window for the insight. If the result of action 840 is no, action 844 may allow observation of the insight to continue (and process 800 may proceed to evaluate the next insight in order of priority). If the result of action 840 is yes, action 846 may update the reinforcement learning parameters with negative feedback regarding previously performed actions for this insight (and process 800 may proceed to evaluate the next insight in order of priority).

[0145] If the result of action 836 is yes, then in action 848, improvement praise can be given. Improvement praise may take into account settings 850, such as one or more styles or methods for coaching. After action 848, the reinforcement learning parameters may be updated in action 852 with positive feedback on the actions previously taken for this insight (and process 800 may proceed to evaluate the next insight in order of priority).

[0146] Figure 9 shows an example of multimodal insight triggering 900. Multimodal insight triggering 900 can be used in combination with one or more other examples described elsewhere in this specification. Here, multimodal insight triggering 900 relates to insight 902. For example, insight 902 may relate to high wake-up glucose and whether to increase the dose of long-acting insulin. Insight 902 can be triggered based on two or more criteria 904 that are met. Each criterion may specify one or more situations that can reflect a performance metric. In some implementation forms, N criteria 904 are used, where N = 1, 2, 3, ...

[0147] Multimodal insight triggering 900 can be multifaceted and may involve two or more performance metrics, including but not limited to behavioral metrics, physiological metrics, and / or therapeutic metrics. Examples of criterion 904 include, but are not limited to, the following: PWD having more than a specified number of days of sustained-acting insulin per week (e.g., behavioral metric), PWD having less than a specified amount of time below range (e.g., physiological metric), PWD having more than a specified amount of coverage in glucose data (e.g., behavioral metric), PWD having fewer than a specified number of treated nocturnal hypoglucose episodes per month, PWD having less than a specified amount of time of nocturnal hypoglucose, PWD having less than a specified median minimum wakefulness glucose level, or PWD having less than a specified daily ratio of sustained-acting insulin dose to rapid-acting insulin dose (e.g., therapeutic metric). Accordingly, action 810 may consider one or more of criterion 904 in determining whether insight 902 is triggered.

[0148] Figure 10 shows an example of a multimodal insight satisfaction rating 1000. The multimodal insight satisfaction rating 1000 can be used in combination with one or more other examples described elsewhere in this specification. Here, the multimodal insight satisfaction rating 1000 relates to insight 902 of the example described above. Insight 902 may have two or more criteria 1002 to satisfy. Each of the criteria 1002 may specify one or more situations that can reflect performance metrics. In some implementations, M criteria 1002 are used, where M = 1, 2, 3, ... The number of satisfaction criteria M may be the same as the number of triggering criteria N, or may be a larger or smaller number.

[0149] The multimodal insight satisfaction rating 1000 can be multifaceted and may involve two or more performance metrics, including but not limited to behavioral metrics, physiological metrics, and therapeutic metrics. Examples of criterion 1002 include, but are not limited to, the following: PWD is greater than a predetermined number of days of sustained-acting dose per week (e.g., behavioral metric); PWD has coverage exceeding a predetermined amount in glucose data (e.g., behavioral metric); PWD has a probability of more than a predetermined number of insulin therapy changes; or PWD is less than a predetermined minimum median wakefulness glucose level. Accordingly, action 836 may consider one or more of criterion 1002 in determining whether the improvement criteria for insight 902 have been met. That is, satisfaction (or dissatisfaction) with criterion 1002 determines whether reinforcement learning generates positive (or negative) feedback for the actions taken with respect to insight 902.

[0150] Figure 11 shows an example 1100 of an action 1102 related to an insight. Action 1102 can be used in combination with one or more other examples described elsewhere in this specification. Action 1102 can be presented on a display device 1104 (for example, within an app on a mobile electronic device). A PWD can opt in (or opt out) of any of the actions 1102. For example, an opt-in input and / or opt-out input can be received. Action 1102 may be presented to encourage a PWD to set goals for managing their diabetes therapy.

[0151] Figure 12 shows an exemplary architecture of a computing device 1200 that can be used to implement aspects of this disclosure, including any of the systems, devices, and / or technologies described herein, or any other systems, devices, and / or technologies that may be used in various anticipated embodiments.

[0152] The computing device shown in Figure 12 can be used to run the operating systems, application programs, and / or software modules (including software engines) described herein. The computing device may utilize at least one non-temporary computer-readable storage medium.

[0153] In some embodiments, the computing device 1200 includes at least one processing device 1202 (e.g., a processor), such as a central processing unit (CPU). Various processing devices are available from various manufacturers, e.g., Intel or Advanced Micro Devices. In this example, the computing device 1200 also includes system memory 1204 and a system bus 1206 that connects various system components, including the system memory 1204, to the processing device 1202. The system bus 1206 is one of any number of bus structures that may be used, including but not limited to a memory bus or memory controller, a peripheral bus, and a local bus using any of various bus architectures.

[0154] Examples of computing devices that can be implemented using computing device 1200 include desktop computers, laptop computers, tablet computers, mobile computing devices (such as smartphones, touchpad mobile digital devices, or other mobile devices), or other devices configured to process digital instructions.

[0155] The system memory 1204 includes read-only memory 1208 and random access memory 1210. For example, during startup, a basic input / output system 1212 containing basic routines that function to transfer information within the computing device 1200 may be stored in the read-only memory 1208.

[0156] In some embodiments, the computing device 1200 may also include a secondary storage device 1214, such as a hard disk drive, for storing digital data. The secondary storage device 1214 is connected to the system bus 1206 by a secondary storage interface 1216. The secondary storage device 1214 and its associated computer-readable medium provide a non-volatile and non-temporary storage device for computer-readable instructions (including application programs and program modules), data structures, and other data for the computing device 1200.

[0157] The exemplary environments described herein employ a hard disk drive as the secondary storage device, but other embodiments may use other types of computer-readable storage media. Examples of these other types of computer-readable storage media include magnetic cassettes, flash memory cards, digital video discs, Bernoulli cartridges, compact disc read-only memory, digital multipurpose disc read-only memory, random access memory, or read-only memory. Some embodiments include non-temporary media. For example, a computer program product may be tangibly embodied in a non-temporary storage medium. In addition, such computer-readable storage media may include local storage devices or cloud-based storage devices.

[0158] Several program modules, including an operating system 1218, one or more application programs 1220, other program modules 1222 (such as software engines as described herein), and program data 1224, can be stored in a secondary storage device 1214 and / or system memory 1204. The computing device 1200 may utilize any suitable operating system, such as Microsoft Windows®, Google Chrome® OS, Apple OS, Unix or Linux and its variations, and any other operating system suitable for the computing device. Other examples include operating systems from Microsoft, Google, or Apple, or any other suitable operating systems used in tablet computing devices.

[0159] In some embodiments, the user provides input to the computing device 1200 via one or more input devices 1226. Examples of input devices 1226 include a keyboard 1228, a mouse 1230, a microphone 1232 (e.g., for voice and / or other audio input), a touch sensor 1234 (such as a touchpad or touch-sensitive display), and a gesture sensor 1235 (e.g., for gesture input). In some implementations, the input devices 1226 provide detection based on presence, proximity, and / or movement. In some implementations, the user may walk into their home, which may trigger input to the processing device. For example, the input device 1226 then facilitates an automated experience for the user. Other embodiments include other input devices 1226. The input devices can be connected to the processing device 1202 via an input / output interface 1236 coupled to the system bus 1206. These input devices 1226 can be connected by any number of input / output interfaces, such as parallel ports, serial ports, game ports, or universal serial buses. Wireless communication between the input devices 1226 and the input / output interfaces 1236 is also possible, and in some possible embodiments, to name a few examples, includes infrared, Bluetooth® wireless technology, 802.11a / b / g / n, cellular, ultra-wideband (UWB), ZigBee, or other radio frequency communication systems.

[0160] In this exemplary embodiment, a display device 1238, such as a monitor, liquid crystal display device, light-emitting diode display device, projector, or touch-sensitive display device, is also connected to the system bus 1206 via an interface such as a video adapter 1240. In addition to the display device 1238, the computing device 1200 may include various other peripheral devices (not shown), such as speakers or printers.

[0161] The computing device 1200 can connect to one or more networks via the network interface 1242. The network interface 1242 can provide wired and / or wireless communication. In some implementations, the network interface 1242 may include one or more antennas for transmitting and / or receiving wireless signals. When used in a local area networking environment or a wide area networking environment (such as the Internet), the network interface 1242 may include an Ethernet interface. Other possible embodiments use other communication devices. For example, some embodiments of the computing device 1200 include a modem for communication across the network.

[0162] The computing device 1200 may include at least some form of computer-readable medium. The computer-readable medium includes any available medium accessible by the computing device 1200. For example, the computer-readable medium includes computer-readable storage media and computer-readable communication media.

[0163] Computer-readable storage media include volatile and non-volatile, removable and non-removable media configured to store information such as computer-readable instructions, data structures, program modules, or other data, and implemented within any device. Examples of computer-readable storage media include, but are not limited to, random-access memory, read-only memory, electrically erasable and writable read-only memory, flash memory or other memory technologies, compact disk read-only memory, digital multipurpose disk or other optical storage devices, magnetic cassettes, magnetic tapes, magnetic disk storage devices or other magnetic storage devices, or any other media that can be used to store desired information and are accessible by computing device 1200.

[0164] Computer-readable communication media typically include any information transmission medium that embodies computer-readable instructions, data structures, program modules, or other data in modulated data signals, such as carrier waves or other transmission mechanisms. The term “modulated data signal” refers to a signal in which one or more of its characteristics are set or modified in a manner that encodes the information in the signal. Examples of computer-readable communication media include wired media such as wired networks or direct wired connections, as well as wireless media such as acoustic, radio frequency, infrared, and other wireless media. Any combination of the above is also included in the scope of computer-readable media.

[0165] The computing device shown in Figure 12 is also an example of a programmable electronic device that may include one or more such computing devices, and if multiple computing devices are included, such computing devices can be coupled together into a suitable data communication network to collectively perform various functions, methods, or operations disclosed herein.

[0166] Exemplary Therapy Management System As mentioned above, a permanent therapy that provides consistent glycemic control is necessary to keep blood glucose levels consistently within acceptable limits. Such glycemic control can be achieved by lowering elevated blood glucose levels by regularly supplying the PWD body with external drugs. External bioeffective drugs (e.g., insulin or its analogues) are generally administered by daily injections. In some cases, multiple daily injections of a mixture of RA insulin and LA insulin are administered via a reusable transdermal fluid delivery device.

[0167] Currently, monitoring A1C levels is limited to every 3 to 6 months, depending on various factors. This method of measuring patient progression and follow-up, limited to quarterly or semi-annual visits, can contribute to significant inertia in therapy throughout PWD treatment, resulting in suboptimal outcomes and a "treat until failure" approach to diabetes care. Furthermore, limited human resources continue to perpetuate delays in adhering to treatment standards.

[0168] Embodiments of therapy management devices, systems, and methods described below can regularly maintain one or more of a user's glucose metrics within a target level or range, provide frequent therapy interventions and adjustments (e.g., daily or weekly, rather than every three to six months), reduce intervention duration, and increase user adherence, outcomes, and satisfaction. System components may include, but are not limited to, a blood glucose meter or monitor, an insulin pump, an injection pen, a smart cap on the injection pen, an insulin therapy software application (app), and / or cloud infrastructure.

[0169] Figure 13 shows the therapy management system 1300 in various exemplary embodiments. The therapy management system 1300 can be configured to provide a therapy escalation pathway for managing diabetes and maintaining one or more glucose metrics to or within the range of one or more targets. The therapy management system 1300 can be further configured to regularly monitor one or more metrics of the user (e.g., A1C levels). The therapy management system 1300 can be further configured to provide frequent therapy interventions and therapy adjustments (e.g., daily or weekly, rather than every 3-6 months). The therapy management system 1300 can be further configured to shorten the intervention period. The therapy management system 1300 can be further configured to increase user adherence, user outcomes, and user satisfaction. The therapy management system 1300 can be further configured to tailor the treatment plan to the user (e.g., based on the user's needs and values). The therapy management system 1300 can be further configured to select possible treatment plans (e.g., dosing regimens) based on past successful interventions with respect to the user or population (e.g., a history of similar treatments with the user). The therapy management system 1300 can be further configured to provide users and / or healthcare professionals with more information more frequently so that more options are available for therapies. The therapy management system 1300 can be further configured to continuously and / or periodically monitor one or more user metrics (e.g., mean blood glucose, weight, BMI, nausea, etc.) so that therapy interventions can be timely and small. The therapy management system 1300 can be further configured to continuously monitor and improve user adherence through coaching and direct feedback (e.g., notifications, warnings, recommendations, requests for user action, etc.) so that lack of treatment adherence and lack of therapy effectiveness can be addressed separately.

[0170] Although the therapy management system 1300 is shown in Figure 13 as a standalone device and / or system, aspects of the present disclosure can be used with other devices, systems, and / or methods, for example, but not limited to, elements in Figures 14A-14C, 15-18, 19A-19C, and 20-23, such as control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, one or more dosing regimens, 1410, 1440, 1450, 1460, 1480, flow diagram 2000, flow diagram 2100, flow diagram 2200, and / or computing device 2300.

[0171] As shown in Figure 13, the therapy management system 1300 may include a remote device 1310, an on-body unit (OBU) 1350, one or more drug delivery devices such as a first injection pen 1360 and a second injection pen 1370, a first smart cap 1382, a second smart cap 1384, and / or a software application 1390. In some embodiments, the therapy management system 1300 may include one or more memories (e.g., memory 1394) and at least one processor (e.g., processor 1392), each processor being coupled to at least one of the memories and configured to run a software application (app) (e.g., software application 1390) configured to monitor one or more user metrics and implement one or more therapy pathways (e.g., one or more dosing regimens) for the user. In some embodiments, the therapy management system 1300 may include drug delivery devices (e.g., injection pens, insulin pumps, etc.) configured to communicate with the software application 1390 and monitor the user and deliver one or more medications (e.g., insulin) to the user. In some embodiments, the therapy management system 1300 may include a scale configured to communicate with a software application 1390 and measure the user's weight.

[0172] In some embodiments, the therapy management system 1300 can be configured to execute one or more therapy escalation pathways. In some embodiments, the therapy management system 1300 can receive user glucose data from a CGM and calculate one or more metrics, and the processor can compare one or more metrics with one or more pre-set or user-adjustable thresholds stored in memory and output recommendations to the user for an adjusted dose and / or new therapy on the computing device's display. In some embodiments, the therapy management system 1300 can determine whether the user is actually taking the recommended dose and / or new therapy, determine one or more new metrics after the user has taken the recommended dose, and recommend an adjusted (titrated) dose based on glucose data. In some embodiments, the therapy management system 1300 can implement a titration phase, in which the therapy management system 1300 requests the user to actively administer the recommended dose, reviews one or more user metrics after the user has taken the medication, and then recommends a new dose based on the user's glucose data.

[0173] The remote device 1310 can be configured to receive or retrieve sensor data from the sample sensor 1352. As shown in Figure 13, the remote device 1310 may include a display 1312. The display 1312 may be a touchscreen display for receiving input. The remote device 1310 may include one or more input devices for receiving input, such as one or more buttons, keys, etc. The display 1312 may present various information to the user, such as glucose data, drug data, alerts, notifications, and recommendations, among other things, to request information from the user, for example. The remote device 1310 may include a bolus calculator accessible via a bolus calculator button 1314. The display may show one or more notifications or recommendations, such as sample level 1321, point indicator 1322, historical sample data 1323, trend display 1324, first message 1331, second message 1332, third message 1333, first information display 1334, and second information display 1335. In some embodiments, the remote device 1310 may include a processor 1392 coupled to a memory 1394 and configured to run a software application 1390. In some embodiments, the remote device 1310 may include a mobile application (e.g., software application 1390) configured to run one or more operations of the therapy management system 1300 (e.g., control diagram 1400A, therapy assessment 1400B, control diagram 1400C).

[0174] As shown in Figure 13, the remote device 1310 can communicate with the OBU 1350 by, for example, one or more wireless communication protocols (e.g., NFC, BLE, etc.). The remote device 1310 can communicate with one or more drug delivery devices or dose detection devices, such as the first injection pen 1360, the second injection pen 1370, the first smart cap 1382, the second smart cap 1384, and / or the software application 1390. In some embodiments, the remote device 1310 may include a handheld computer (e.g., a smartphone, cell phone, mobile phone, PDA, smartwatch, etc.), a personal computer, a laptop computer, a dedicated handheld device associated with a sample sensor or drug delivery device, or any other portable communication device.

[0175] As shown in Figure 13, in an exemplary display 1312 of an exemplary mobile application (e.g., software application 1390) running on a dedicated smartphone, the display 1312 may include a bolus calculator button 1314 and, optionally, one or more other buttons or inputs for the user to enter data or to respond to recommendations and / or requests. In some embodiments, the display 1312 may include a numerical and / or graphical display of glucose data (e.g., historical, current, and / or predicted). For example, as shown in Figure 13, the display 1312 may include a glucose level 1321 showing an estimate of the current (or most recent) glucose level, a point indicator 1322 graphing the current (or most recent) estimated glucose level as part of an overall graph, historical sample data 1323 showing a graph of historical glucose data (e.g., the last 30 minutes), and a trend display 1324 showing a predicted trend or expected continuation of the historical glucose data 1323.

[0176] In some embodiments, the display 1312 may include one or more text messages to the user regarding the current therapy. For example, as shown in Figure 13, the display 1312 may include a first message 1331 (e.g., a notification), a second message 1332 (e.g., a recommendation), and a third message 1333 (e.g., a request for user action). In some embodiments, the first message 1331 may include a notification and / or warning to the user describing, for example, one or more characteristics of glucose data (e.g., "high glucose trend"). In some embodiments, the second message 1332 may include a recommendation for therapy to the user, for example, a recommendation for one or more treatments or therapies (e.g., "you probably need more insulin"). In some embodiments, the second message 1332 may provide recommendations for insulin, carbohydrates, and / or other therapies. In some embodiments, the third message 1333 may include a request for user action, for example, a request to the user to respond to or perform a specific action (e.g., "take 15g of carbohydrates").

[0177] In some embodiments, the display 1312 may allow the user to click on glucose data (e.g., blood glucose level 1321) or navigate within a mobile app (e.g., software application 1390) to obtain additional, more detailed, or more complete glucose data. In some embodiments, the display 1312 may provide one or more insights (e.g., communications, notifications, recommendations, requests, etc.) for the benefit of the user (e.g., PWD). For example, an insight could deliver a therapy-related, tailored message to the user: "You normally take a meal bolus between noon and 2 p.m., but you did not take a bolus today, and your blood glucose level is currently elevated. Did you forget your bolus?" In some embodiments, display 1312 may display insulin data and / or administration data. In some embodiments, display 1312 may include a first information display 1334 configured to display information for one or more administration regimens, including but not limited to the time and / or amount of the most recent dose of RA and / or LA insulin. In some embodiments, display 1312 may include a second information display 1335 configured to display information for one or more insulin onboard (IOB) amounts, including but not limited to the amount of IOB for a particular type of insulin or insulin analogue (e.g., RA, LA, etc.). In some embodiments, IOB (e.g., active insulin) may be defined as the amount of insulin delivered that remains active in the user's body based on the estimated duration of insulin action. In some embodiments, the IOB calculation may be for both RA and LA insulin. In some embodiments, the display 1312 may allow the user to click on insulin data (e.g., the first information display 1334) or navigate within a mobile app (e.g., the software application 1390) to obtain additional, more detailed, or more complete insulin delivery data. In some embodiments, the display 1312 may overlay glucose data and insulin delivery data in any preferred format, such as a graphical display of glucose data timing versus insulin delivery data timing.

[0178] In some embodiments, during use, a user (e.g., a PWD and / or healthcare professional) can utilize the therapy management system 1300 to receive recommendations regarding appropriate insulin dosages and therapies. For example, if there is an immediate need to deliver LA insulin, a second message 1332 may provide a recommendation for a specific LA insulin dosage. In some embodiments, the recommended dosage from the therapy management system 1300 (e.g., a software application 1390) may be displayed on the displays of the first smart cap 1382 and / or the second smart cap 1384, for example, on the first display 1383 and / or the second display 1385, respectively.

[0179] In some embodiments, the user may enter a bolus calculator using the bolus calculator button 1314 to calculate and deliver a bolus of RA insulin, for example. In some embodiments, the bolus calculator button 1314 may include any suitable bolus calculator, for example, the bolus calculator may provide a user interface for the user to input a meal announcement as either correction only, small meal, normal-sized meal, or large meal. Once the meal size or meal type (e.g., breakfast, lunch, dinner) has been selected, the user interface may provide a recommended bolus dose, for example, based on the number of carbohydrates associated with the corresponding meal size and / or meal type, and optionally based on glucose data.

[0180] The OBU 1350 can be configured to measure and communicate sensor data from one or more user samples (e.g., glucose). The OBU 1350 can be further configured to communicate data from the sample sensor 1352 (e.g., sensor data) to one or more other components of the therapy management system 1300 (e.g., remote device 1310, first injection pen 1360, second injection pen 1370, first smart cap 1382, second smart cap 1384, software application 1390, etc.). In some embodiments, the OBU 1350 can exchange data with a network and / or remote server. As shown in Figure 13, the OBU 1350 may include a sample sensor 1352, on-body electronics 1354, on-body housing 1355, and / or adhesive layer 1356.

[0181] The sample sensor 1352 can be configured to measure a user's sample (e.g., glucose). The sample sensor 1352 can be configured to measure one or more of a user's samples (e.g., glucose, ketones). The sample sensor 1352 can be further configured to measure the concentration of one or more of a user's samples in real time and continuously (e.g., in vivo). In some embodiments, a portion of the sample sensor 1352 (e.g., tip portion) can be placed in vivo through the patient's skin surface (e.g., transdermally) and in fluid contact with the patient's bodily fluids (e.g., interstitial fluid). In some embodiments, the sample sensor 1352 may be insertable into the patient's body containing the sample (e.g., vein, artery, skin, etc.). In some embodiments, the sample sensor 1352 may include a continuous glucose monitor (CGM) for continuous and automatic tracking of blood glucose levels. In some embodiments, the sample sensor 1352 can measure and retrieve sample levels in real time (e.g., approximately 1 to 60 seconds) for continuous sample (glucose) monitoring.

[0182] In some embodiments, the sample sensor 1352 can measure glucose. In some embodiments, the sample sensor 1352 can measure lactate. In some embodiments, the sample sensor 1352 can measure ketones. In some embodiments, the sample sensor 1352 can measure one or more samples. For example, the sample sensor 1352 can measure glucose and ketones. For example, the sample sensor 1352 can detect one or more samples (e.g., glucose, ketone bodies, lactate, lactic acid, oxygen, hemoglobin A1C, lactone, lactose, galactose, vitamin C, glucuronic acid, glycogen, mannose, phosphate, bisphosphate, fructose, glyceraldehyde, glycerol, triglycerides, sorbitol, phosphogluconate, phosphogluconic acid, xylulose, ribose, bile, cysteine, serine, homoserine, pyruvate, phenylpyruvic acid, glutamic acid, glycine, taurine, threonine, methionine, ethanol, acetone, acetate, oxaloacetate, alanine, phenylalanine, aspartic acid, asparagine, alcohol, cholesterol, vitamin D, pro It can measure (gesterone, testosterone, estrogen, squalene, insulin, hydroxybutyrate, leucine, isoleucine, malonyl, malonic acid, glucagon, epinephrine, norepinephrine, palmitate, lysine, eicosanoids, melanin, dopamine, tyrosine, tryptophan, niacin, melatonin, serotonin, citric acid, isocitrate, valine, porphyrin, histidine, urocanate, histamine, glutamine, proline, creatine, putrescine, spermidine, spermine, arginine, ornithine, citrulline, fumarate, succinate, argininosuccinate, succinyl, ketoglutaric acid, aconitic acid, glyoxylic acid, caffeine, sugars, carbohydrates, etc.). In some embodiments, the sample sensor 1352 can simultaneously measure one or more samples using one or more corresponding electrochemical biosensors for each of the different samples to be measured.

[0183] The on-body electronic device 1354 can be configured to process signals from the sample sensor 1352. The on-body electronic device 1354 can be further configured to communicate data from the sample sensor 1352 (e.g., sensor data) to one or more external devices (e.g., remote device 1310, first injection pen 1360, second injection pen 1370, first smart cap 1382, second smart cap 1384, software application 1390, etc.). The on-body electronic device 1354 can be further configured to wirelessly transmit sensor data related to the sample (e.g., NFC, WiFi, Bluetooth, BLE, internet, etc.). As shown in Figure 13, the on-body electronic device 1354 can be operably (e.g., electrically) coupled to the sample sensor 1352 and can be wirelessly coupled to the remote device 1310, the first injection pen 1360, the second injection pen 1370, the first smart cap 1382, the second smart cap 1384, and / or the software application 1390.

[0184] In some embodiments, the on-body electronics 1354 may include a printed circuit board (PCB) for connection to various components (e.g., sample sensor 1352, processor, ASIC, wireless transceiver, wireless transmitter, controller, memory, etc.). In some embodiments, the on-body electronics 1354 may store historical sample-related data (e.g., via memory). In some embodiments, the on-body electronics 1354 may be configured to store some or all of the sample-related data (e.g., sensor data) from the sample sensor 1352 in memory. In some embodiments, the on-body electronics 1354 may include one or more processors and / or control logic configured to determine (e.g., via software programs and / or algorithms) the current sample level, the rate of change (ROC) of the sample level, the rate of acceleration of the sample level, and / or sample trend information (e.g., trend display 1324), and / or sample variability levels (e.g., standard deviation, variability, variability, etc.).

[0185] In some embodiments, the on-body electronic device 1354 can be configured to transmit (broadcast) specimen-related data (e.g., sensor data) to one or more external devices (e.g., remote device 1310, first injection pen 1360, second injection pen 1370, first smart cap 1382, second smart cap 1384, software application 1390, etc.). In some embodiments, the on-body electronic device 1354 can be configured to transmit (broadcast) real-time data from the specimen sensor 1352, associated with the monitored specimen level, to one or more of its external devices in the therapy management system 1300, for example, when the external devices are within the communication range (e.g., BLE range) of data broadcast from the OBU 1350.

[0186] In some embodiments, the on-body electronic device 1354 can be configured to wirelessly transmit specimen-related sensor data stored during a monitoring period (e.g., sensor attachment) to one or more external devices of the therapy management system 1300. In some embodiments, specimen-related data (e.g., sensor data) transmitted from the on-body electronic device 1354 can be stored (e.g., permanently, temporarily) in one or more memory units, for example, memory units on one or more external devices of the therapy management system 1300. In some embodiments, the remote device 1310 can be configured as a data conduit for passing data (e.g., sensor data) received from the on-body electronic device 1354 to one or more external devices. In some embodiments, the on-body electronic device 1354 can be designed to store sensor data (e.g., glucose data) from the specimen sensor 1352 collected over a sensor attachment period (e.g., 3 days, 7 days, 14 days, 30 days, etc.), for example, 14 days.

[0187] The on-body housing 1355 can be configured to provide internal compartments for a portion of the specimen sensor 1352 (e.g., the proximal portion) and the on-body electronics 1354. As shown in Figure 13, the on-body housing 1355 may include the specimen sensor 1352 and the on-body electronics 1354 and can be bonded to the adhesive layer 1356. In some embodiments, the on-body housing 1355 may include a sealed housing (e.g., a hermetically sealed biocompatible housing). The adhesive layer 1356 can be configured to attach the OBU 1350 to the user's skin surface. As shown in Figure 13, the adhesive layer 1356 can be bonded to the on-body housing 1355 to ensure that a portion of the specimen sensor 1352 (e.g., the tip portion) is securely positioned on the skin surface.

[0188] The first injection pen 1360 can be configured to administer one or more dosing regimens. In some embodiments, the first injection pen 1360 can be combined with a first smart cap 1382 to transfer drug dose data to a remote device 1310, an OBU 1350, and / or a software application 1390. As shown in Figure 13, the first injection pen 1360 may include a first dial 1361 configured to set the dose to be delivered, and a first dose indicator 1362 configured to show the dose. In some embodiments, the first injection pen 1360 may include a corresponding first smart cap 1382 that communicates wirelessly with other components of the therapy management system 1300. In some embodiments, the first injection pen 1360 may include an insulin injection pen for a specific type of insulin, such as a rapid-acting (RA) insulin pen. For example, the first injection pen 1360 may include a Humalog® pen, a Novolog® pen, or an Apidra® pen. In some embodiments, the first injection pen 1360 may include dose capture technology to automatically capture, for example, injection data (e.g., set dose, injected dose, dose history, date, administration time, time since last administration, drug type, amount of drug remaining in the pen, etc.) via the first smart cap 1382.

[0189] The second injection pen 1370 can be configured to administer one or more dosing regimens. In some embodiments, the second injection pen 1370 can be combined with a second smart cap 1384 to transfer drug dose data to a remote device 1310, an OBU 1350, and / or a software application 1390. As shown in Figure 13, the second injection pen 1370 may include a second dial 1371 configured to set the dose to be delivered, and a second dose indicator 1372 configured to show the dose. In some embodiments, the second injection pen 1370 may include a corresponding second smart cap 1384 that communicates wirelessly with other components of the therapy management system 1300. In some embodiments, the second injection pen 1370 may include an insulin injection pen for a specific type of insulin, such as a long-acting (LA) insulin pen. For example, the second injection pen 1370 may include a Lantus® pen, a Levemir® pen, a Toujeo® pen, or a Tresiba® pen. In some embodiments, the second injection pen 1370 may include dose capture technology for automatically capturing injection data (e.g., dose, dose history, date, time, etc.) via, for example, a second smart cap 1384.

[0190] The first smart cap 1382 can be configured to communicate wirelessly with a remote device 1310, an OBU 1350, and / or a software application 1390. The first smart cap 1382 can be further configured to present information to the user, including but not limited to any information or sensor data described elsewhere in this specification (e.g., insulin dosage recommendations, notifications, communications, etc.). In some embodiments, the first smart cap 1382 can exchange data with the OBU 1350 (e.g., retrieve sensor data from the OBU 1350) by, for example, bringing it into close proximity. In some embodiments, the first smart cap 1382 can receive meal sizes from the user, including but not limited to small, medium, large, or snack sizes (e.g., the user announces a meal size, selects a meal size, etc.). In some embodiments, the first smart cap 1382 can include at least one input control, such as a button (e.g., an electronic switch). Input control (e.g., a button) can be configured to trigger the first smart cap 1382 to perform one or more actions, including but not limited to any actions described elsewhere in this specification. As shown in Figure 13, the first smart cap 1382 may include the first display 1383.

[0191] The first display 1383 may include any preferred type of display technology, including but not limited to dynamic electronic displays (e.g., LED displays) or static electronic displays (e.g., e-ink displays). In some embodiments, the first display 1383 may include a user interface for obtaining dietary bolus recommendations for different types of meals (e.g., small, medium, large meals; breakfast, lunch, dinner, snacks; 10 grams of carbohydrates, 20 grams of carbohydrates, 45 grams of carbohydrates, etc.). In some embodiments, the first display 1383 may display a first message 1331, a second message 1332, a third message 1333, a first information display 1334, a second information display 1335, or a combination thereof.

[0192] The second smart cap 1384 can be configured to communicate wirelessly with the remote device 1310, the OBU 1350, and / or the software application 1390. In some embodiments, the second smart cap 1384 can exchange data with the OBU 1350 (e.g., retrieve sensor data from the OBU 1350) by, for example, bringing it into close proximity to the OBU 1350. In some embodiments, the second smart cap 1384 can receive a meal size from the user, including but not limited to small, medium, large, or snack (e.g., the user announces the meal size, selects the meal size, etc.). In some embodiments, the second smart cap 1384 can include at least one input control, e.g., a button (e.g., an electronic switch). The input control (e.g., a button) can be configured to trigger the second smart cap 1384 to perform one or more actions, including but not limited to any actions described elsewhere in this specification. As shown in Figure 13, the second smart cap 1384 may include a second display 1385.

[0193] The second display 1385 may include any preferred type of display technology, including but not limited to dynamic electronic displays (e.g., LED displays) or static electronic displays (e.g., e-ink displays). In some embodiments, the second display 1385 may include a user interface for obtaining dietary bolus recommendations for different types of meals (e.g., small, medium, large meals; breakfast, lunch, dinner, snacks; 10 grams of carbohydrates, 20 grams of carbohydrates, 45 grams of carbohydrates, etc.). In some embodiments, the second display 1385 may display a first message 1331, a second message 1332, a third message 1333, a first information display 1334, a second information display 1335, or a combination thereof.

[0194] In some embodiments, the first smart cap 1382 and / or the second smart cap 1384 may be configured to record, store, and / or transmit one or more types of pen capping information, for example, information about when the first smart cap 1382 and / or the second smart cap 1384, respectively, were attached to and / or released from the corresponding first injection pen 1370 and / or the second injection pen 1360. In some embodiments, the pen capping information may include information about the current capping period (e.g., time since the last capping), information about the duration of one or more uncappings (which may also be referred to herein as “decappings”), and / or the timing of each uncapping and each capping (e.g., time or elapsed time). For example, the pen capping information may include data reflecting when the first smart cap 1382 and / or the second smart cap 1384 were placed on the corresponding first injection pen 1360 and / or the second injection pen 1360, respectively, when the first smart cap 1382 and / or the second smart cap 1384 were removed from the corresponding first injection pen 1370 and / or the second injection pen 1370, respectively, or both. In some embodiments, the pen capping information may be presented to the user on the displays of the first smart cap 1382 and / or the second smart cap 1384 (for example, via the first display 1383 and / or the second display 1385). In some embodiments, the first smart cap 1382 and / or the second smart cap 1384 may include a speaker, microphone, receiver, or other audio device, and the pen capping information may be detected by a receiver (e.g., a microphone) in the first smart cap 1382 and / or the second smart cap 1384 and / or presented by a speaker.

[0195] In some embodiments, for example, the first smart cap 1382 and / or the second smart cap 1384 may provide a timer clock that counts up from the last time fixed to the first injection pen 1360 and / or the second injection pen 1370, respectively. In some embodiments, the first smart cap 1382 and / or the second smart cap 1384 may wirelessly transmit pen capping information to one or more components of the therapy management system 1300 (e.g., remote device 1310) and / or remote computing devices (e.g., remote server, remote computer, network, cloud server, smartphone, tablet, etc.). In some embodiments, one or more accessories or smart delivery devices may detect other events associated with the drug delivery action of the therapy management system 1300 and use that information in the manner described herein with respect to pen capping information. For example, in some cases, an injection pen accessory (e.g., a smart cap) may be attached to the injection pen, which can then detect the mechanical movement of the administration mechanism and determine the drug administration time, and, optionally, the amount of drug delivered at that time.

[0196] In some embodiments, pen-capping information can be stored, displayed, and / or analyzed in combination with glucose data to determine user behavior, such as whether the user is administering insulin appropriately for meals, and / or to correct elevated blood glucose levels. In some embodiments, pen-capping information can be presented on a graphical representation of the user's glucose data and presented to the user and / or healthcare professionals. In some embodiments, glucose data from the period following each capping event of the first smart cap 1382 and / or the second smart cap 1384 can be evaluated to determine whether the user administered insulin appropriately for that capping event, such as appropriate dose, underdose, overdose, basal overdose, etc.

[0197] In some embodiments, a pen-capping event may be ignored if other information indicates that no dose was delivered. For example, an event may be ignored if no change in dose selection was detected for the first injection pen 1360 and / or the second injection pen 1370 (e.g., the first dial 1361 and / or the second dial 1371, respectively). In some embodiments, pen uncapping and recapping events may be ignored if the total uncapping time is less than a first threshold (e.g., about 4-6 seconds). For example, the first threshold may be determined based on a time amount that is too short to allow an injection but long enough to allow the user to check the end of the pen to see if there is any insulin remaining or if there is a needle attached to the pen. In some embodiments, the total decapping time for a decapping event (e.g., the time between the uncapping event and the subsequent recapping) may be analyzed in combination with glucose data to determine whether an injection occurred during that decapping event. In some embodiments, if the total decapping time exceeds a second threshold (e.g., at least 15 minutes, at least 30 minutes, etc.), the approximate time of injection can be determined using the methods described herein for detecting glucose data and meal timing.

[0198] In some embodiments, some or all components of the therapy management system 1300 can perform one or more actions or functions as described elsewhere in this specification. In some embodiments, the therapy management system 1300 can evaluate insights based on reinforcement learning and perform one or more actions (e.g., one or more dosing regimens) in relation to one or more triggered insights. In some embodiments, the therapy management system 1300 can determine insulin use by detecting the removal and / or replacement of the first smart cap 1382 and / or the second smart cap 1384. In some embodiments, the therapy management system 1300 can detect the amount of insulin and / or other medications administered (e.g., GLP-1, prandial insulin, or MDI of prandial insulin).

[0199] The software application 1390 can be configured to monitor one or more user metrics and implement one or more therapeutic pathways (e.g., one or more dosing regimens) for the user. The software application 1390 can be further configured to provide therapeutic escalation pathways for managing diabetes and maintaining one or more glucose metrics within a target level or range. The software application 1390 can be further configured to guide users with diabetes in managing their insulin therapy. The software application 1390 can be further configured to provide frequent therapeutic interventions and adjustments (e.g., daily or weekly, rather than every three to six months). The software application 1390 can be further configured to continuously and / or periodically monitor one or more user metrics (e.g., mean blood glucose, one or more glucose metrics, weight, BMI, nausea, etc.) so that therapeutic interventions can be timely and small. The software application 1390 can be further configured to continuously monitor and improve user adherence through coaching and direct feedback (e.g., notifications, warnings, recommendations, requests for user action) so that lack of treatment adherence and lack of therapeutic effect can be addressed separately. As shown in Figure 13, the remote device 1310 may include a processor 1392 coupled to memory 1394 and configured to run the software application 1390.

[0200] As shown in Figure 13, the software application 1390 can be executed by one or more processors (e.g., processors, controllers, microprocessors, microcontrollers, ASICs, etc.), for example, a processor 1392 on a remote device 1310. In some embodiments, the software application 1390 can be coupled to a memory for storing instructions, for example, a memory 1394 on a remote device 1310, and the instructions, when executed, cause one or more processors to perform operations including, but not limited to, recommending a first dosing regimen of basal insulin, determining basal excess of the first dosing regimen, recommending the addition of a second dosing regimen of GLP-1 or dual GIP / GLP-1, recommending the addition of a third dosing regimen of prandial insulin for each meal, recommending the addition of a fourth dosing regimen of graded MDI of prandial insulin, and recommending the addition of a fifth dosing regimen of basal insulin and prandial insulin for each meal.

[0201] In some embodiments, the software application 1390 may be part of the remote device 1310. In some embodiments, the software application 1390 may be part of the OBU 1350, the first smart injection pen 1360, the second smart injection pen 1370, the first smart cap 1382, and / or the second smart cap 1384. In some embodiments, the software application 1390 may include a mobile application (app). In some embodiments, the software application 1390 may include a web application (app). For example, the software application 1390 may be part of a remote device (e.g., within a web app) located in a remote location (e.g., a medical professional's computer, a service center, etc.).

[0202] In some embodiments, all processing and functionality required for the software application 1390 may be included within the mobile app. In some embodiments, some or part of the software application 1390 (e.g., the mobile app) may be included in a remote server supporting the software application 1390, for example, the remote server may support the mobile app by having processing, communication hub, and / or reporting functions. In some embodiments, the functions for the software application 1390 described herein include functions in the mobile app, the remote server, or both, and it should be noted that all or some functions may be in either or both. In some embodiments, references to the software application 1390 described herein mean both the mobile app and the web server (e.g., the remote server) supporting the mobile app.

[0203] In some embodiments, the software application 1390 (e.g., a mobile app) can acquire continuous glucose sensor data in real time. In some embodiments, the software application 1390 can acquire various combinations of individual sample measurements and continuous sample measurements.

[0204] In some embodiments, the software application 1390 can be configured to evaluate one or more insights based on reinforcement learning and to determine whether to generate each of the insights. In some embodiments, each insight can be associated with performing one or more actions. For example, performing an action may include providing a message to the user and / or automatically changing insulin therapy. In some embodiments, the insights can be tailored to cover the user's entire diabetes therapy, including but not limited to diabetes management, maintaining one or more glucose metrics at or within a target level, weight loss, one or more dosing regimens, and / or one or more therapy escalation pathways. In some embodiments, the insights may include detecting observations from data (e.g., sensor data, user metrics, etc.) so that specific therapy actions can be taken. In some embodiments, the insights are sufficiently relevant to the overall therapy so that the user's therapy (e.g., MDI therapy) can be significantly improved.

[0205] In some embodiments, the software application 1390 can be configured to implement a therapy assessment of the current dosing regimen to monitor one or more user compliance metrics over a first period (e.g., a 3-day cycle) and to confirm user adherence to the therapy. In some embodiments, the software application 1390 can be configured to implement a therapy assessment to rapidly titrate basal insulin over a second period (e.g., a 3-day cycle). In some embodiments, rapid titration may include recommending an adjusted dose over a first interval (e.g., daily) based on one or more user metrics over that first interval until no change is observed over a predetermined period (e.g., 3 days). In some embodiments, the software application 1390 can be configured to implement a therapy assessment to monitor for no change in the dosing output state over a third period (e.g., the first four consecutive no-change outputs). In some embodiments, the software application 1390 can be configured to implement a therapy assessment to maintain basal insulin titration over a fourth period (e.g., a 14-day cycle). In some aspects, maintaining titration may include recommending an adjusted dose over a second interval (e.g., several days, one week, two weeks) based on one or more user metrics for that second interval, until a new issue (e.g., a new medication regimen, a new exercise routine, a new diet, etc.) is identified.

[0206] In some embodiments, the software application 1390 can be configured to determine basal excess based on a comparison of one or more metrics with a predetermined threshold. The threshold may be pre-set by the system or adjustable by the user or healthcare professional. In some embodiments, basal excess may be determined based on the ratio of TDD to body weight. For example, the threshold may be based on a total insulin dose exceeding 0.5 unit(U) / kg / day. Total daily dose (TDD) may be determined by summing the total amount of insulin taken by the user in a day. In some embodiments, the TDD may be entered manually by the user. In some embodiments, the user may manually enter each insulin dose administered, and the system may calculate the TDD by summing the individual doses. In some embodiments, doses may be automatically recorded by a drug delivery device used to administer insulin, such as an insulin pump, a smart insulin pen, or a dose monitoring device attached to a drug delivery device, such as a smart pen cap. Such a device may communicate dose information to the system for TDD calculation. A combination of manually entered doses and automatically taken doses may be used to determine the TDD. Weight can be manually entered by the user into a computing device. In some embodiments, the scale can communicate with the system. The user may weigh themselves using the scale, and the scale may automatically communicate the measured weight to the system. The system may periodically prompt the user to weigh themselves, for example, at corresponding intervals in the rapid titration phase or the sustained titration phase.

[0207] In some embodiments, the software application 1390 can be configured to determine basal excess based at least in part on the glucose difference between waking time (AM) and bedtime (PM) and / or the glucose difference after prandial and before prandial, or the rate of change in blood glucose levels overnight or during meals, or a combination thereof.

[0208] In some embodiments, the software application 1390 can be configured to determine basal excess based at least in part on hypoglycemia metrics exceeding a threshold. The system may use glucose data to determine hypoglycemia metrics such as time-below range (TBR). The system may be configured to output a glucose alarm to the user if the user's blood glucose level falls below a low glucose threshold or below a very low glucose threshold. The system may track the number of hypoglycemia alarm outputs or very low alarm outputs during a predetermined period. Hypoglycemia may be determined if the number of alarms during the predetermined period exceeds a threshold.

[0209] In some embodiments, the software application 1390 can be configured to determine basal excess based at least in part on an increase in glucose variability. Basal excess may be detected when glucose variability exceeds a predetermined variability threshold. Glucose variability may be calculated using a coefficient of variation (e.g., the ratio of the standard deviation to the mean). In some embodiments, the glucose variability threshold may include a coefficient of variation greater than 25%, 30%, or 35%, for example. In some embodiments, glucose variability may be determined based, among other things, on an alternative measure of variability, such as the standard deviation from the mean or the interquartile range.

[0210] In some embodiments, the software application 1390 can be configured to implement one or more dosing regimens based on one or more targets of the user. In some embodiments, for example, the targets can be based at least in part on one or more metrics of the user. In some embodiments, for example, the targets can be based at least in part on the user's BMI and the user's mean blood glucose level. In some embodiments, if one or more of the user's parameters exceed one or more of the user's targets, the software application 1390 can be configured to escalate the therapy to keep one or more parameters within or to the range of one or more targets by recommending the addition and / or modification of one or more dosing regimens.

[0211] In some embodiments, the therapy management system 1300 may include a processor configured to run a software application (app) 1390 configured to monitor one or more user metrics and implement one or more therapy pathways (e.g., one or more dosing regimens) for the user. In some embodiments, the software application 1390 may be part of a remote device 1310, an OBU 1350, a first injection pen 1360, a second injection pen 1370, a first smart cap 1382, and / or a second smart cap 1384. In some embodiments, the software application 1390 may include a mobile application on the remote device 1310. In some embodiments, the therapy management system 1300 may include a network or remote server configured to support the software application 1390. In some embodiments, the software application 1390 may be entirely contained within a user mobile application (app), or some or part of the software application 1390 may be contained within a remote server (e.g., a web server, cloud server, intranet server, etc.) that supports a software application having, for example, processing, communication, and / or reporting functions. In some embodiments, the software application 1390 may include one or more application programming interfaces (APIs) for two or more computer programs to communicate with each other.

[0212] In some embodiments, the software application 1390 can be configured to recommend one or more therapy escalation pathways. In some embodiments, the software application 1390 can receive user glucose data from a CGM (e.g., OBU 1350), calculate one or more user metrics (e.g., as described herein), compare one or more user metrics to one or more pre-configured or user-adjustable targets (thresholds) stored in memory, and output recommendations to the user for adjusted doses and / or new therapies to the display 1312 of the remote device 1310. In some embodiments, the software application 1390 can determine whether the user is actually taking the recommended doses and / or new therapies (e.g., monitoring one or more glucose metrics), determine one or more new metrics after the user has taken the recommended doses, and recommend adjusted (titrated) doses based on glucose data. In some embodiments, the software application 1390 may recommend a titration phase, during which the software application 1390 may request the user to actively administer the recommended dose, review one or more user metrics after the user has taken the drug, and then recommend a new dose based on the user's glucose data.

[0213] In some embodiments, the software application 1390 can be configured to determine the number of days for one or more dosing regimens (e.g., basal insulin regimens). For example, the number of days can be defined by the time between LA doses. In some embodiments, the software application 1390 can define the number of days for basal titration based on the time between LA doses. In some embodiments, for example, short days where the time between LA doses is less than a first interval (e.g., about 18 hours) can be considered invalid and discarded for determining the number of days for basal titration. In some embodiments, for example, short days where the time between LA doses exceeds a second interval (e.g., about 18 hours) can be considered valid and used for determining the number of days for basal titration. In some embodiments, for example, long days where the time between LA doses exceeds a third interval (e.g., about 24 hours) can be considered valid and used for determining the number of days for basal titration. In some embodiments, for example, long days where the time between LA doses exceeds a fourth interval (e.g., about 24 hours) can be considered good days (e.g., 24-hour periods) and used for determining the number of days for basal titration. In some embodiments, for example, a long day between LA doses exceeding a fifth interval (e.g., approximately 42 hours) can be considered a missed dose day (e.g., 42 hours since the last dose) and can be used to determine the number of days of basal titration. In some embodiments, missed dose days can be repeated over a sixth interval (e.g., every 24 hours) until another LA dose is taken and can be used to determine the number of days of basal titration.

[0214] Example control diagram Figure 14A shows control diagram 1400A for implementing one or more therapy pathways in various exemplary embodiments. Control diagram 1400C can be configured to provide a therapy escalation pathway for managing diabetes. Control diagram 1400C can be further configured to implement one or more therapy pathways (e.g., one or more dosing regimens) for the user. Control diagram 1400C can be further configured to assess the appropriateness of each dosing regimen and evaluate for basal excess.

[0215] Although control diagram 1400A is shown in Figure 14A as a standalone process, apparatus, and / or system, aspects of the present disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14B, 14C, 15-18, 19A-19C, and 20-23, such as the therapy management system 1300, software application 1390, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1450, 1460, 1480, flow diagram 2000, flow diagram 2100, flow diagram 2200, and / or computing device 2300.

[0216] As shown in Figure 14A, the control diagram 1400A may include a first therapy 1401, a therapy assessment 1403, and a second therapy 1405. In some embodiments, the control diagram 1400A may be implemented by a software application 1390. In some embodiments, the control diagram 1400A may be implemented for one or more dosing regimens (e.g., one or more dosing regimens 1410, 1440, 1450, 1460, 1480) to provide one or more therapy escalation pathways. In some embodiments, the control diagram 1400A may include a first dosing regimen 1410, a second dosing regimen 1440, a third dosing regimen 1450, a fourth dosing regimen 1460, a fifth dosing regimen 1480, or a combination thereof.

[0217] The first therapy 1401 can be configured to recommend the first therapy when one or more user metrics are outside the target range. In some embodiments, the first therapy 1401 may include a first dosing regimen 1410 (e.g., basal insulin), a second dosing regimen 1440 (e.g., GLP-1 or dual GIP / GLP-1), or a combination thereof. In some embodiments, one or more user metrics may include one or more glucose metrics (e.g., mean glucose, median glucose, glucose TIR, glucose TBR, glucose TAR, glucose TVL, GMI, MMG, MPDG, PPG, beam, or a combination thereof), BBR, ISF, PRA-CG, body weight, BMI, TDD (e.g., insulin), TDD-to-body weight ratio, A1C level (e.g., glycated hemoglobin percentage, mmol / mL), heart rate, blood pressure, or a combination thereof. In some embodiments, one or more targets may include a target blood glucose level (e.g., approximately 110 mg / dL), a target blood glucose range (e.g., approximately 110 mg / dL to approximately 155 mg / dL), a target basal state (e.g., MMG approximately 100 mg / dL to approximately 130 mg / dL), a target diet state (e.g., MPDG approximately 100 mg / dL to approximately 120 mg / dL), a target modified ISF state (e.g., PRA-CG approximately 70 mg / dL to approximately 180 mg / dL), a target body weight, a target A1C level (e.g., 5.7%), a target heart rate, a target blood pressure, or a combination thereof. In some embodiments, one or more targets may be based at least in part on the user's BMI and mean blood glucose level.

[0218] Therapy assessment 1403 can be configured to assess the appropriateness of the first therapy 1401 and evaluate for basal excess. Therapy assessment 1403 can be further configured to confirm user adherence to the first therapy 1401. In some embodiments, therapy assessment 1403 can be configured to determine basal excess based at least in part on drug delivery information. For example, basal excess may be determined at least in part on the ratio of insulin TDD to the user's body weight above a predetermined threshold (e.g., above about 0.5 U / kg / day). In some embodiments, therapy assessment 1403 can be configured to determine basal excess based at least in part on one or more glucose metrics. For example, basal excess can be determined at least partially based on changes in blood glucose levels overnight (e.g., the difference between waking time (AM) and bedtime (PM)), changes in blood glucose levels during a meal (e.g., the difference between pre-prandial and post-prandial), hypoglycemic metrics that do not exceed a predetermined threshold (e.g., a glucose concentration of approximately 70 mg / dL), glucose fluctuations that exceed a predetermined threshold (e.g., a pre-meal rise of approximately 30 mg / dL, or a post-meal spike exceeding approximately 110 mg / dL), or a combination of these.

[0219] The second therapy 1403 may be configured to recommend adding the second therapy if one or more user metrics remain outside one or more targets, or if basal excess is determined. In some embodiments, the second therapy 1403 may include a second dosing regimen 1440 (e.g., GLP-1 or dual GIP / GLP-1), a third dosing regimen 1450 (e.g., prandial insulin per meal), a fourth dosing regimen 1460 (e.g., graded MDI of prandial insulin), a fifth dosing regimen 1480 (e.g., basal insulin and prandial insulin per meal), or a combination thereof.

[0220] Figure 14B shows a therapeutic assessment 1400B in an exemplary embodiment. The therapeutic assessment 1400B can be configured to assess the appropriateness of the current therapy, verify user adherence, and recommend adjusting one or more doses based on one or more user metrics. The therapeutic assessment 1400B can be further configured to be implemented for any of the dosing regimens described herein (e.g., one or more dosing regimens 1410, 1440, 1450, 1460, 1480).

[0221] Therapy Assessment 1400B is shown in Figure 14B as a standalone process, apparatus, and / or system, but aspects of the present disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14A, 14C, 15-18, 19A-19C, and 20-23, such as the Therapy Management System 1300, software application 1390, control diagram 1400A, control diagram 1400C, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1450, 1460, 1480, flow diagram 2000, flow diagram 2100, flow diagram 2200, and / or computing device 2300.

[0222] As shown in Figure 14B, the therapy assessment 1400B may include a monitoring phase 1402, a rapid titration phase 1404, and a maintenance titration phase 1406. In some embodiments, the therapy assessment 1400B can be implemented by a software application 1390. In some embodiments, the therapy assessment 1400B can be implemented for one or more dosing regimens (e.g., one or more dosing regimens 1410, 1440, 1450, 1460, 1480) to assess the current therapy and adjust the dose based on one or more user metrics.

[0223] Monitoring phase 1402 can be configured to monitor one or more user compliance metrics of the current therapy over a predetermined period. Monitoring phase 1402 can be further configured to determine whether the user is properly wearing and using the CGM (e.g., OBU 1350) and administering the medication appropriately and consistently. In some embodiments, one or more user compliance metrics may include the frequency of glucose data (scans), the time between gaps in glucose data, the time of administration, the amount of dose (e.g., relative to the recommended dose), the receipt of user input, or a combination thereof. In some embodiments, the predetermined period may be in the range of about 1 to about 7 days, for example, 3 days. In some embodiments, once one or more user compliance metrics are received (met) over the predetermined period (e.g., 3 days), the current therapy can proceed to rapid titration phase 1404.

[0224] The rapid titration phase 1404 can be configured to recommend an adjusted dose for a first interval (e.g., daily midnight) based on one or more user metrics (e.g., listed above) during that first interval. In some embodiments, the rapid titration phase 1404 may continue the process until no change is observed over a predetermined period (e.g., 3 days out of 7, 6 days, etc.). In some embodiments, dose adjustments may be at regular intervals (e.g., an increase of 0.25 U, a 10% increase, etc.) or proportional to the degree of glycemic dysfunction (e.g., a larger dose change if significantly deviating from the glucose target). In some embodiments, if no change is received over a predetermined period (e.g., 3 days out of 7, 6 days, etc.), the current therapy may proceed to maintenance titration phase 1406.

[0225] The maintenance titration phase 1406 can be configured to recommend an adjusted dose for a second interval (e.g., several days, one week) based on one or more user metrics (e.g., those listed above) during that second interval. In some embodiments, the maintenance titration phase 1406 may continue the process until a new problem with the current therapy is identified. In some embodiments, dose adjustments may be at fixed intervals (e.g., an increase of 0.25 U, a 10% increase, etc.) or proportional to the degree of glycemic dysfunction (e.g., a larger dose change if the glucose target is significantly off). In some embodiments, the new problem may include a new medication, a new exercise routine, a new diet, or a combination thereof.

[0226] Exemplary dosing regimen Figures 14C, 15–18, and 19A–19C show control diagram 1400C for implementing one or more dosing regimens 1410, 1440, 1450, 1460, and 1480 in various exemplary embodiments. Control diagram 1400C can be configured to provide a therapy escalation pathway for managing diabetes. Control diagram 1400C can be further configured to implement one or more therapy pathways (e.g., one or more dosing regimens) for the user. Control diagram 1400C can be further configured to assess the validity of each dosing regimen and evaluate for basal excess.

[0227] Although control diagram 1400C is shown in Figure 14C as a standalone process, apparatus, and / or system, aspects of the present disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14A, 14B, 15-18, 19A-19C, and 20-23, such as the therapy management system 1300, software application 1390, control diagram 1400A, therapy assessment 1400B, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1450, 1460, 1480, flow diagram 2000, flow diagram 2100, flow diagram 2200, and / or computing device 2300.

[0228] As shown in Figure 14C, control diagram 1400C may include a first dosing regimen 1410 (e.g., basal insulin), a therapy assessment 1430, a second dosing regimen 1440 (e.g., GLP-1 or dual GIP / GLP-1), a third dosing regimen 1450 (e.g., prandial insulin per meal), a fourth dosing regimen 1460 (e.g., graded MDI of prandial insulin), and / or a fifth dosing regimen 1480 (e.g., basal insulin and prandial insulin per meal). In some embodiments, control diagram 1400C may implement one or more therapies for the user, including one or more of the first dosing regimen 1410, a therapy assessment 1430, a second dosing regimen 1440, a third dosing regimen 1450, a fourth dosing regimen 1460, a fifth dosing regimen 1480, or a combination thereof. In some embodiments, the control diagram 1400C can be implemented by a software application 1390.

[0229] In some embodiments, for example, the software application 1390 may monitor user data (e.g., one or more user metrics) of the therapy management system 1300 (e.g., via the OBU 1350, the first injection pen 1360, the second injection pen 1370, the first smart cap 1382, and / or the second smart cap 1384) and send one or more control signals or commands to the therapy management system 1300 (e.g., via the remote device 1310, the first smart cap 1382, and / or the second smart cap 1384) to execute or command the user to execute one or more dosing regimens of control diagram 1400C and evaluate whether an excessive basal excess has occurred for one or more dosing regimens.

[0230] The first dosing regimen 1410 can be configured to initiate basal insulin therapy. In some embodiments, if one or more user metrics are outside one or more targets (e.g., mean glucose, median glucose, glucose TIR, glucose TBR, glucose TVL, GMI, MMG, PPG, BBR, BeAM, ISF, body weight, BMI, A1C level, heart rate, blood pressure, or a combination thereof), the software application 1390 can implement the first dosing regimen 1410. In some embodiments, one or more targets may include a target basal state (e.g., MMG between approximately 100 mg / dL and approximately 130 mg / dL). In some embodiments, the first dosing regimen 1410 may be based at least in part on user-specific considerations, for example, the selection of basal insulin may be based on cost and any existing conditions of the user. In some embodiments, the first dosing regimen 1410 may be based at least in part on the prescription of glucagon for hypoglycemia that has occurred. In some embodiments, the first dosing regimen 1410 may have an initial starting dose of basal insulin of approximately 0.1 U / kg / day to approximately 0.2 U / kg / day. In some embodiments, the first dosing regimen 1410 may include adding a continuous glucose monitor (CGM) (e.g., OBU 1350) to the user to set an initial glucose target (e.g., the glucose target may be in the range of approximately 117 mg / dL to approximately 154 mg / dL).

[0231] In some embodiments, the first dosing regimen 1410 may include a monitoring phase to ensure that the user is using the CGM appropriately and adhering to the therapy regimen. This may be determined by monitoring one or more of the following: system use, sensor placement, sensor scan or completeness of glucose data, and the dose administered by the user (correspondence between the administered dose and the recommended dose). For example, the monitoring phase may be of a first period, which may be one day or longer. One or more compliance metrics may be assessed daily, and therapy adherence may be determined daily based on the compliance metrics. If a compliance threshold is reached, such as meeting the compliance target for three consecutive days, the user may be considered to be adhering to the therapy. In some embodiments, if the user is not adhering to the first dosing regimen 1410, the software application 1390 may provide coaching and interventions to help the user adhere to the first dosing regimen 1410 more effectively. For example, coaching may include alarms and reminders (e.g., messages reminding the user to take their therapy), education or encouragement (e.g., messages advising the user on the importance of regular therapy use, and encouragement to continue using the therapy), and / or long-term education (e.g., general in-person or virtual diabetes education to better understand what is happening in the user's body and the value of using the therapy). Notifications may be based on compliance metrics. For example, if the compliance metric relates to insulin dose timing and the user is not taking insulin doses at consistent times, a notification may alert the user and provide a recommendation to take the dose at a specific time. If the compliance metric indicates incomplete glucose data (e.g., less than 70% of daily glucose data), a notification may warn the user to ensure that the sensor is worn and the remote device is kept within range, or to scan the sensor more frequently to obtain glucose data.

[0232] In some forms, coaching can be directed at the user. For example, coaching may include short educational messages (e.g., "Do you know that behavior A is associated with outcome Y?", "Why not try walking for 45 minutes today?", "Do you know that walking enables message Z?"), enrollment in adherence alarms or alerts (e.g., missed dose alert, low alert, get moving alert), motivational messages to encourage behavior (e.g., "You scanned your CGM at least four times every day last week, keep it up!", "You took drug A, and over the past month, you took drug A on time 95% of the time. You said you felt better on the days you didn't."), and / or long-term education (e.g., attending a cooking class, enrolling in a virtual type 2 diabetes education class).

[0233] In some aspects, interventions may be directed towards healthcare professionals. For example, interventions may include encouraging users to take their medication daily (e.g., not needing to escalate therapy), considering easier-to-administer medications (e.g., once-daily medication, oral or injectable medication), considering strengthening current therapy (e.g., physiological marker patterns do not match expected behavior, insulin is not lowering blood glucose, GLP-1 is not reducing weight), considering additional therapy (e.g., basal insulin → basal insulin + GLP-1) because users are taking medication but not seeing therapeutic benefits, and / or considering alternative therapies (e.g., basal insulin → basal insulin + GLP-1).

[0234] In some embodiments, the software application 1390 can define preferred therapy algorithms (e.g., control diagrams 1400A and 1400C). For example, the preferred therapy algorithm can be defined by the user (e.g., user values ​​and goals) or by one or more healthcare professionals. In some embodiments, for example, the preferred therapy algorithm can include a predetermined therapy escalation pathway (e.g., diet and exercise → metformin → metformin + GLP-1 → GLP-1 + basal insulin → GLP-1 + MDI). In some embodiments, for example, the preferred therapy algorithm can exclude certain medications and / or certain therapy pathways from consideration (e.g., not using SGLT-2 agents or Flozin, not using diet and exercise, etc.). In some embodiments, the software application 1390 can allow healthcare professionals to specify what changes the user can directly make to the preferred therapy algorithm (e.g., the user can increase the insulin dose by 20%, the user can increase the GLP-1 dose by one step (10 μg over two weeks), etc.).

[0235] In some embodiments, if a user is adhering to a first dosing regimen 1410 but is not achieving therapeutic effects, the software application 1390 can provide one or more options for adjusting the therapy (e.g., the route of therapy). For example, for a user using an injectable drug, the software application 1390 can adjust the dosage (e.g., basal insulin dose). For example, for a user using an oral drug, the software application 1390 can send a message to the user's HCP to adjust the dosage of the oral drug and / or to request the addition of another oral drug to the user's current therapy. For example, the software application 1390 can escalate the user's therapy to a higher-intensity drug (e.g., basal insulin → MDI, metformin → GLP-1, etc.). In some embodiments, if a user is adhering to a first dosing regimen 1410 but is not achieving therapeutic effects, the software application 1390 can implement a second dosing regimen 1440 (e.g., basal insulin → basal insulin + GLP-1).

[0236] Therapeutic assessment 1430 can be configured to evaluate the appropriateness of one or more dosing regimens and to assess basal excess. Therapeutic assessment 1430 can be further configured to determine basal excess by comparing one or more metrics with one or more thresholds for those metrics. In some embodiments, therapeutic assessment 1430 can be performed for a first dosing regimen 1410. In some embodiments, therapeutic assessment 1430 can determine basal excess based at least in part on the ratio of the total daily dose (TDD) of insulin to the user's body weight. In some embodiments, for example, the ratio of insulin TDD to body weight can be compared to a predetermined threshold (e.g., about 0.5 U / kg / day). In some embodiments, therapeutic assessment 1430 can determine basal excess based at least in part on glucose fluctuations. In some embodiments, for example, glucose fluctuations can be compared to one or more predetermined thresholds (e.g., a pre-meal rise of about 30 mg / dL, a post-meal spike of about 110 mg / dL, or the total change in blood glucose from the pre-meal period to the post-meal period). In some embodiments, the therapy assessment 1430 can determine basal excess based at least in part on a hypoglycemic metric. In some embodiments, for example, the hypoglycemic metric may include, among other things, a hypoglycemic index (LBGI), time-below range (TBR), the number of low glucose alarms, and the number of very low glucose alarms, among other hypoglycemic indicators. In some embodiments, for example, the hypoglycemic metric may be compared to a predetermined threshold (e.g., a glucose concentration below approximately 70 mg / dL). In some embodiments, the therapy assessment 1430 can determine basal excess based at least in part on the change in blood glucose levels during a meal. In some embodiments, for example, the change in blood glucose levels during a meal may be compared to a predetermined threshold (e.g., the difference between post-prandial and pre-prandial). In some embodiments, the therapy assessment 1430 can determine basal excess based at least in part on the change in blood glucose levels overnight. In some embodiments, for example, the change in blood glucose levels overnight may be compared to a predetermined threshold (e.g., the difference between waking (AM) and going to bed (PM)).

[0237] If a basal excess is detected, the software application 1390 may be configured to recommend that the user initiate a second dosing regimen. The remote device 1310 may output a notification on the display 1312 to provide a recommendation to initiate a second dosing regimen. Alternatively, or in addition, the system may provide a notification to the HCP's computing device to recommend initiating a second dosing regimen. In some embodiments, the computing device may generate a report containing the recommended dose regimen.

[0238] A second dosing regimen 1440 may include the addition of GLP-1 to an existing therapy, such as basal insulin therapy. GLP-1 may be added if it is not contraindicated for the patient and is not already part of the user's therapy regimen. The system may retrieve current therapy information to determine whether the therapy regimen already includes GLP-1. The system may include contraindication information, which may be entered by the user or HCP to determine whether GLP-1 is contraindicated.

[0239] In some embodiments, at the initiation of GLP-1 receptor agonist (RA) therapy, basal insulin undergoes a predetermined initial reduction. For example, the current basal insulin dose may be reduced by only 20% of the current basal insulin dose. The basal insulin dose may be further adjusted by reducing or increasing the basal insulin dose during the rapid titration phase, as described herein.

[0240] GLP-1 RAs may be titrated concurrently with or subsequently with basal insulin or other existing therapies. GLP-1 RAs may be titrated within the approved range of a particular GLP-1 RA. GLP-1 RAs may be initiated at a first dose and increased escalating toward the maximum dose. The dose may be increased escalating, for example, on a unit basis or a partial unit basis, as long as the GLP-1 RA is tolerated by the user.

[0241] In some embodiments, the system may recommend increasing the dose of GLP-1 RA if no adverse events such as nausea are reported or detected. The system may prompt the user for input regarding adverse events. The system may specifically ask if the user is experiencing nausea or vomiting. The system may ask the user how they are feeling in free-form text. The system may ask the user to select from a predetermined list of adverse events. In some embodiments, the system may be configured to automatically detect nausea (e.g., via an external sensor). If nausea is reported, the dose of GLP-1 RA may be maintained at the current level and further increases are not recommended. Instead, if nausea is reported, the dose of GLP-1 RA may be reverted to a previous dose of GLP-1 RA administered that the user reported did not cause nausea.

[0242] In some embodiments, GLP-1 RAs are also titrated based on the user's weight. The system may determine the user's target weight. The target weight may be entered by the user, based on input from, for example, a health care provider (HCP), or entered directly by the HCP, or may be a default setting. The target weight may be based on BMI. The user's weight may be determined by a smart scale. The smart scale may automatically transmit the weight measurement to the system. The system may prompt the user to weigh themselves. The user may be prompted to weigh themselves. The user may be prompted regularly, at intervals of, among others, every two days, every three days, or once a week. Alternatively, the system may prompt the user to weigh themselves if no weight measurement has been received for a given period of time. If the user's weight falls below the target weight, further increases to the GLP-1 RA may not be recommended to avoid excessive weight loss. Instead, if the user's weight falls below the target weight, the dose of GLP-1 RA may be reduced to curb further weight loss.

[0243] The second dosing regimen 1440 can be configured to add GLP-1 receptor agonist therapy or dual GIP / GLP-1 receptor agonist therapy. In some embodiments, if one or more user metrics exceed one or more targets (e.g., mean glucose, median glucose, glucose TIR, glucose TBR, glucose TVL, GMI, MMG, PPG, BBR, BeAM, ISF, body weight, BMI, A1C level, heart rate, blood pressure, or a combination thereof), the software application 1390 can implement the second dosing regimen 1440. In some embodiments, one or more targets may include a target basal state (e.g., MMG of approximately 100 mg / dL to approximately 130 mg / dL), target body weight, target dietary state (e.g., minimum post-administration glucose (MPDG) of approximately 100 mg / dL to approximately 120 mg / dL), target corrected ISF state (e.g., rapid-acting post-administration CG with corrected glucose of approximately 70 mg / dL to approximately 180 mg / dL), or a combination thereof. In some embodiments, a second administration regimen 1440 may include a GLP-1 receptor agonist or a dual GIP / GLP-1 receptor agonist in any combination with insulin (e.g., basal insulin) or in a fixed ratio combination. In some embodiments, a second administration regimen 1440 (e.g., GLP-1 or dual GIP / GLP-1) may be added to the first administration regimen 1410 (e.g., basal insulin). For example, if one or more user metrics remain above one or more targets after the first dosing regimen 1410, or if basal excess is determined for the first dosing regimen 1410, the software application 1390 may recommend adding a second dosing regimen 1440 to the first dosing regimen 1410. In some embodiments, it may be recommended to add a second dosing regimen 1440 (e.g., GLP-1 or dual GIP / GLP-1) to the first dosing regimen 1410 (e.g., basal insulin).For example, if one or more user metrics remain above one or more targets after the first dosing regimen 1410, or if basal excess is determined for the first dosing regimen 1410, the software application 1390 may recommend adding a second dosing regimen 1440 to the first dosing regimen 1410.

[0244] If basal excess is detected and the user's therapy regimen is already fully titrated and includes GLP-1 RA, or if GLP-1 RA is contraindicated, the system determines whether the pre- and post-meal glucose difference is large for a single meal in the user. The pre- and post-meal glucose difference can be based on the change in blood glucose levels from the blood glucose level at the start of the meal or the time before the meal to the maximum blood glucose level after the meal or the maximum blood glucose level during a predetermined period after the meal (e.g., 2-5 hours).

[0245] A third dosing regimen 1450 can be configured to add prandial insulin therapy for a single meal. In some embodiments, if the postprandial glucose elevation for a single meal exceeds a predetermined threshold, the software application 1390 can implement the third dosing regimen 1450. In some embodiments, the predetermined threshold can be defined by the total change in blood glucose levels from the pre-meal period to the post-meal period (difference between minimum and maximum values), the rate of change in blood glucose levels (slope), the difference between post-prandial and pre-prandial, or a combination thereof. In some embodiments, prandial insulin therapy can include bolus insulin, RA insulin, "meal-time" insulin, aspart, glulisine, lispro, or a combination thereof. In some embodiments, the third dosing regimen 1450 (e.g., prandial insulin for a single meal) can include a single dose of prandial insulin for the user's largest meal. In some embodiments, the third dosing regimen 1450 (e.g., prandial insulin per meal) may include a single dose of prandial insulin per meal for the user with the greatest PPG fluctuation. In some embodiments, it may be recommended to add the third dosing regimen 1450 (e.g., prandial insulin per meal) to the second dosing regimen 1440 (e.g., GLP-1 or dual GIP / GLP-1). For example, if the postprandial glucose elevation for a meal exceeds a predetermined threshold after the second dosing regimen 1440, or if basal excess is determined for the second dosing regimen 1440, the software application 1390 may recommend adding the third dosing regimen 1450 to the second dosing regimen 1440.

[0246] In some embodiments, the third dosing regimen 1450 may have an initial starting dose of prandial insulin per meal, based on a predetermined amount or a ratio or percentage of prandial insulin to basal insulin. For example, the initial dose of prandial insulin may be about 4 U / day, or about 10% of the once-daily basal insulin dose. In some embodiments, if the user's A1C level is less than 8% (e.g., about 64 mmol / mol), the third dosing regimen 1450 may recommend reducing the basal insulin dose by only about 4 U / day, or about 10% of the once-daily basal insulin dose. In some embodiments, with respect to hypoglycemia, the software application 1390 may determine the cause of hypoglycemia by analyzing one or more user metrics, for example, based on hypoglycemia indices, particularly the low blood glucose index (LBGI), time-below range (TBR), number of low glucose alarms, and number of very low glucose alarms. In some embodiments, if there is no clear reason for hypoglycemia (for example, based on the metrics above), the once-daily insulin dose can be reduced by approximately 10% to 20% for safety reasons.

[0247] In some embodiments, the plandial dose is added to meals that cause glucose fluctuations, such as the meal with the greatest pre- and post-meal glucose increase. In some embodiments, the plandial dose is added to the largest meal of the day.

[0248] The fourth dosing regimen 1460 can be configured to add a stepwise MDI of prandial insulin therapy to the corresponding meals. If the system determines, based on glucose data, that there is a large pre-meal and post-meal glucose difference for two or more meals, the system may recommend adding multiple daily injections (MDIs). Instead of adding a prandial insulin dose for a single meal, the prandial insulin dose may be added for two meals or for each of the three main meals of the day (i.e., breakfast, lunch, and dinner). In some embodiments, the MDI is the final therapeutic escalation recommended by the system.

[0249] In some embodiments, one or more targets may include a target basal state (e.g., MMG of approximately 100 mg / dL to approximately 130 mg / dL and BBR < 1.5), a target meal state (e.g., MPDG of approximately 100 mg / dL to approximately 120 mg / dL), a target modified ISF state (e.g., PRA-CG of approximately 70 mg / dL to approximately 180 mg / dL), or a combination thereof. In some embodiments, the fourth dosing regimen 1460 may include a graded MDI of prandial insulin such that the therapy includes two, then three, additional injections daily (e.g., for each meal). In some embodiments, it may be recommended to add the fourth dosing regimen 1460 (e.g., graded MDI of prandial insulin) to the third dosing regimen 1450 (e.g., prandial insulin for one meal). For example, if one or more user metrics remain above one or more targets after the third dosing regimen 1450, the software application 1390 may recommend adding a fourth dosing regimen 1460 to the third dosing regimen 1450.

[0250] The fifth dosing regimen 1480 can be configured to add basal insulin and prandial insulin therapy for each meal. In some embodiments, if one or more user metrics exceed one or more targets (e.g., mean glucose, median glucose, glucose TIR, glucose TBR, glucose TVL, GMI, MMG, PPG, BBR, BeAM, ISF, body weight, BMI, A1C level, heart rate, blood pressure, or a combination thereof), the software application 1390 may recommend adding the fifth dosing regimen 1480. In some embodiments, one or more targets may include a target basal state (e.g., MMG between approximately 100 mg / dL and approximately 130 mg / dL, and BBR < 1.5), a target meal state (e.g., MPDG between approximately 100 mg / dL and approximately 120 mg / dL), a target modified ISF state (e.g., PRA-CG between approximately 70 mg / dL and approximately 180 mg / dL), or a combination thereof. In some embodiments, it may be recommended to add a fifth dosing regimen 1480 (e.g., basal insulin and prandial insulin therapy for each meal) to a fourth dosing regimen 1460 (e.g., a stepwise MDI of prandial insulin). For example, if one or more user metrics remain above one or more targets after the fourth dosing regimen 1460, the software application 1390 may recommend adding the fifth dosing regimen 1480 to the fourth dosing regimen 1460.

[0251] Figure 15 shows a therapeutic assessment 1430 in an exemplary embodiment. Although the therapeutic assessment 1430 is shown in Figure 15 as a standalone process, apparatus, and / or system, embodiments of the present disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14A-14C, 16-18, 19A-19C, and 20-23, such as the therapeutic management system 1300, software application 1390, control diagram 1400A, therapeutic assessment 1400B, control diagram 1400C, one or more dosing regimens 1410, 1440, 1450, 1460, 1480, flow diagram 2000, flow diagram 2100, flow diagram 2200, and / or computing device 2300.

[0252] As shown in Figure 15, the therapy assessment 1430 may include a monitoring phase 1432, a rapid titration phase 1434, a no-change phase 1436, and a maintenance titration phase 1438. In some embodiments, the therapy assessment 1430 may be based at least in part on the ratio of insulin TDD to the user's body weight, glucose variability, hypoglycemia metrics, changes in blood glucose levels during meals or overnight, or a combination thereof. In some embodiments, the therapy assessment 1430 may be based at least in part on the ratio of insulin TDD to the user's body weight, increases in the morning (AM)-bedtime (PM) difference and / or post-prandial-pre-prandial difference exceeding a predetermined threshold (e.g., basal insulin dose exceeding approximately 0.5 U / kg / day), hypoglycemia metrics not exceeding a predetermined threshold, glucose variability exceeding a predetermined threshold (e.g., pre-meal rise exceeding approximately 30 mg / dL and / or post-meal spike exceeding approximately 110 mg / dL), or a combination thereof.

[0253] The monitoring phase 1432 can be configured to monitor one or more user compliance metrics of the current therapy over a predetermined period of time to ensure user adherence to the therapy. In some embodiments, the software application 1390 can monitor one or more of the following: system use, sensor placement, sample sensor scanning, drug dose administration, or a combination thereof, by wirelessly tracking sensor data over time (e.g., via OBU 1350) to ensure that the user is properly fitted with and scanning the sensors (e.g., tracking the number of scans using a remote device 1310, glucose data is collected for a predetermined portion of the day, glucose TBR, glucose TVL, number of low events, etc.); and by wirelessly tracking drug (administration) data over time (e.g., via a drug delivery device, first smart cap 1382) to ensure that the user is properly administered (e.g., tracking weekly dose, time between administrations, dose regularity, etc.).

[0254] In some embodiments, the software application 1390 may implement the monitoring phase 1432 until a compliance threshold is met. The compliance threshold may be based on the meeting of one or more compliance metrics (e.g., as described herein) over a predetermined period, e.g., three consecutive days. In some embodiments, the monitoring phase 1432 may continue to be implemented without allowing a change or escalation of the therapy until user adherence to the current therapy (e.g., meeting the compliance threshold) is observed for a certain period, e.g., three consecutive days.

[0255] Once the monitoring phase is complete, the application proceeds to the rapid titration phase 1434. During the rapid titration phase 1434, the system recommends a drug dose at a first interval, such as once daily. The rapid titration phase 1434 can titrate the dose by fixed increments, e.g., by increasing or decreasing by a fixed number of units, or by a fixed percentage of the current dose. In some embodiments, the rapid titration phase 1434 may be configured to apply a risk-based rapid dose adjustment of the first dosing regimen 1410. In some embodiments, the dose adjustment may be proportional to the deviation of the glucose metric from the corresponding target; for example, a maximum dose adjustment may be recommended when the glucose metric is farther from the target than when the glucose metric is closer to the target. In some embodiments, the user administers a drug dose, and the system can determine one or more user metrics in the period following dose administration to determine whether the user's blood glucose control has improved and whether one or more glucose metrics are within or within the target level. If the glucose metric is not within the target range, the system may recommend increasing or decreasing the drug dose. The system may also monitor one or more drug (dosage) data during rapid titration phase 1434 to titrate the drug within its standard dosage range. The system may generate notifications to the user and / or HCP to adjust the drug dosage.

[0256] In some embodiments, during the rapid titration phase 1434, the software application 1390 may recommend dose adjustments to basal insulin at first intervals, for example, daily or on a 24-hour basis. In some embodiments, the software application 1390 may monitor the user's response to therapy during the rapid titration phase 1434 by determining one or more glucose metrics, as discussed herein, and comparing the metrics to target levels or ranges. For example, the system may wirelessly track sensor glucose data over time (e.g., tracking how far fasting glucose or wake glucose, glucose TAR, glucose TBR, etc., are from the target range) and drug (administration) data over time (e.g., tracking LA insulin administered the previous day, no overnight treatment for hyperglycemia and / or hypoglycemia, sufficient overnight and / or wake data, sufficient time between the last RA administration and wake glucose, etc.). In some embodiments, during the rapid titration phase 1434, on any day when glucose is significantly low, as may be determined based on TBR exceeding the TBR threshold, and if the number of low glucose alarms exceeds a predetermined number, the software application 1390 will recommend reducing the basal insulin dose by a predetermined amount (e.g., by about 10% to about 20%) in an attempt to correct the low glucose.

[0257] In some embodiments, the software application 1390 can implement a rapid titration phase 1434 over a period of time, for example, in a 3-day cycle. In some embodiments, the rapid titration phase 1434 can be implemented for three consecutive days of administration with sufficient glucose (e.g., basal insulin). In some embodiments, the rapid titration phase 1434 may continue to be implemented without allowing any change in therapy or titration until the user demonstrates sufficient glucose control (e.g., glucose metrics at or within the range of one or more targets) for three consecutive days of administration with sufficient glucose over a predetermined period of time.

[0258] The no-change phase 1436 can be configured to monitor the user's response to the first dosing regimen 1410 (e.g., basal insulin). The no-change phase 1436 can be further configured to monitor any side effects of the first dosing regimen 1410. In some embodiments, the software application 1390 can monitor one or more basal states 1436a-1436h and a range from the target basal state (e.g., MMG from approximately 100 mg / dL to approximately 130 mg / dL) during the no-change phase 1436. In some embodiments, the software application 1390 can monitor the user's response to therapy during the no-change phase 1436 by wirelessly tracking sensor data over time (e.g., tracking total glucose TBR, glucose TVL, MMG, BeAM, basal excess indicator, etc.) and by wirelessly tracking drug (dosing) data over time (e.g., tracking treated nocturnal lows, etc.).

[0259] In some embodiments, the software application 1390 can implement the no-change phase 1436 over a period of time, for example, over a no-change cycle. In some embodiments, the no-change phase 1436 can be implemented for a sequence of four consecutive no-change outputs (e.g., basal states 1436a to 1436h). In some embodiments, the user can continue to implement the no-change phase 1436 over a continuous period of time, for example, over the first four consecutive no-change outputs, until no change is observed (e.g., until side effects disappear), without allowing therapeutic changes or escalation.

[0260] As shown in Figure 15, the no-change phase 1436 can be based on a continuous no-change state of one or more basal states 1436a to 1436h. In some embodiments, one or more basal states 1436a to 1436h may include a first basal state 1436a (e.g., significantly below target), a second basal state 1436b (e.g., significantly below target), a third basal state 1436c (e.g., below target), a fourth basal state 1436d (e.g., target), a fifth basal state 1436e (e.g., above target), a sixth basal state 1436f (e.g., significantly above target), a seventh basal state 1436g (e.g., significantly above target), an eighth basal state 1436h (e.g., basal excess), or a combination thereof.

[0261] In some embodiments, the first basal state 1436a can be determined based on a total glucose TBRm greater than approximately 4% or a glucose TVL of less than approximately 1%. In some embodiments, the second basal state 1436b can be determined based on an MMG less than approximately 80 mg / dL or a treated nocturnal low of more than approximately 30 minutes. In some embodiments, the third basal state 1436c can be determined based on an MMG of approximately 80 mg / dL to approximately 100 mg / dL or a BeAM of less than approximately -20 mg / dL / hr. In some embodiments, the fourth basal state 1436d can be determined based on an MMG of approximately 100 mg / dL to approximately 130 mg / dL. In some embodiments, the fifth basal state 1436e can be determined based on an MMG of approximately 130 mg / dL to approximately 150 mg / dL. In some embodiments, the sixth basal state 1436f can be determined based on an MMG of approximately 150 mg / dL to approximately 180 mg / dL. In some embodiments, the seventh basal state 1436g can be determined based on an MMG greater than approximately 180 mg / dL. In some embodiments, the eighth basal state 1436h can be determined by a therapeutic assessment 1430 or by one or more of the above-mentioned basal excess metrics (e.g., TDD to body weight ratio, glucose fluctuation, hypoglycemia metric, glucose variability during meals, overnight glucose variability).

[0262] Once the rapid titration phase is complete, the software application 1390 (e.g., an algorithm) proceeds to the maintenance titration phase 1438. The maintenance titration phase 1438 can be configured to apply the first dosing regimen 1410 (e.g., basal insulin) over a longer period and monitor the user's response to the therapy. The maintenance titration phase 1438 can be further configured to account for significant changes in the user's life (e.g., adaptation to a new diet and / or exercise, addition of a new antidiabetic drug, ongoing changes in insulin sensitivity). In some embodiments, during the maintenance titration phase 1438, the software application 1390 may recommend increasing the dose much more slowly than decreasing it. For example, a day with excessive TBR may result in a recommendation to decrease the daily dose, but the recommended dose increase may require a multi-day pattern of high MMG). In some embodiments, the software application 1390 can monitor the user's response to the therapy during the maintenance titration phase 1438 by determining one or more glucose metrics, as discussed herein, and comparing the metrics to target levels or ranges. For example, the system can wirelessly track sensor glucose data over time (e.g., tracking average glucose, weekly average glucose, glucose TIR, etc.) and drug (administration) data over time (e.g., tracking time between administrations, time between last LA administrations, absence of overnight treatment for hyperglycemia and / or hypoglycemia, sufficient overnight and / or wake-up data, sufficient time between last RA administration and wake-up glucose, etc.).

[0263] In some embodiments, the software application 1390 can implement a maintenance titration phase 1438 over a period of time, for example, over a 14-day cycle. In some embodiments, the maintenance titration phase 1438 can be implemented over 12 days of a 14-day administration with sufficient glucose (e.g., basal insulin). In some embodiments, the maintenance titration phase 1438 may continue to be implemented over a predetermined period, for example, over 12 days of a 14-day administration with sufficient glucose, until the user demonstrates sufficient glucose control (e.g., glucose metrics at or within the range of one or more targets), without allowing any change in therapy or titration.

[0264] Figure 16 shows a second dosing regimen 1440 in an exemplary embodiment. Although the second dosing regimen 1440 is shown in Figure 16 as a standalone process, apparatus, and / or system, embodiments of the present disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14A-14C, 15, 17, 18, 19A-19C, and 20-23, such as the therapy management system 1300, software application 1390, control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, one or more dosing regimens 1410, 1450, 1460, 1480, flow diagram 2000, flow diagram 2100, flow diagram 2200, and / or computing device 2300.

[0265] For example, an embodiment of therapy assessment 1430 shown in Figure 15 and an embodiment of the second dosing regimen 1440 shown in Figure 16 may be similar. Similar reference numerals are used to indicate the features of the embodiment of therapy assessment 1430 shown in Figure 15 and the similar features of the embodiment of the second dosing regimen 1440 shown in Figure 16. One difference between the embodiment of therapy assessment 1430 shown in Figure 15 and the embodiment of the second dosing regimen 1440 shown in Figure 16 is that therapy assessment 1430 has basal insulin, a rapid titration phase 1434, and a no-change phase 1436, while the second dosing regimen 1440 includes a GLP-1 receptor antagonist or a dual GIP / GLP-1 receptor antagonist and a rapid titration phase 1444 (e.g., 6 days of a 7-day administration with sufficient glucose).

[0266] As shown in Figure 16, the second dosing regimen 1440 may include a monitoring phase 1442, a rapid titration phase 1444, and a maintenance titration phase 1446. The monitoring phase 1442 may be configured to monitor one or more user compliance metrics of the current therapy over a predetermined period to ensure user adherence to the therapy. The rapid titration phase 1444 may be configured to introduce and administer a rapid dose of the second dosing regimen 1440 (e.g., both a GLP-1 receptor antagonist or a dual GIP / GLP-1 receptor antagonist and LA insulin) and monitor the user's response to the additional therapy. The maintenance titration phase 1446 may be configured to administer the second dosing regimen 1440 (e.g., both a GLP-1 receptor antagonist or a dual GIP / GLP-1 receptor antagonist and LA insulin) over a longer period and monitor the user's response to the therapy.

[0267] Figure 17 shows a third dosing regimen 1450 in an exemplary embodiment. Although the third dosing regimen 1450 is shown in Figure 17 as a standalone process, apparatus, and / or system, embodiments of the present disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14A-14C, 15, 16, 18, 19A-19C, and 20-23, such as the therapy management system 1300, software application 1390, control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1460, 1480, flow diagram 2000, flow diagram 2100, flow diagram 2200, and / or computing device 2300.

[0268] For example, an embodiment of therapy assessment 1430 shown in Figure 15 and an embodiment of the third dosing regimen 1450 shown in Figure 17 may be similar. Similar reference numerals are used to indicate the features of the embodiment of therapy assessment 1430 shown in Figure 15 and the similar features of the embodiment of the third dosing regimen 1450 shown in Figure 17. One difference between the embodiment of therapy assessment 1430 shown in Figure 15 and the embodiment of the third dosing regimen 1450 shown in Figure 17 is that therapy assessment 1430 has basal insulin and a no-change phase 1436, while the third dosing regimen 1450 includes prandial insulin for one meal.

[0269] As shown in Figure 17, the third dosing regimen 1450 may include a monitoring phase 1452, a rapid titration phase 1454, and a maintenance titration phase 1456. The monitoring phase 1452 may be configured to monitor one or more user compliance metrics of the current therapy over a predetermined period to ensure user adherence to the therapy. The rapid titration phase 1454 may be configured to introduce and apply a rapid dose of the third dosing regimen 1450 (e.g., prandial insulin per meal) and monitor the user's response to the additional therapy. The maintenance titration phase 1456 may be configured to apply the third dosing regimen 1450 (e.g., prandial insulin per meal) over a longer period and monitor the user's response to the therapy.

[0270] Figure 18 shows a fourth dosing regimen 1460 in an exemplary embodiment. Although the fourth dosing regimen 1460 is shown in Figure 18 as a standalone process, apparatus, and / or system, embodiments of the present disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14A-14C, 15-17, 19A-19C, and 20-23, such as the therapy management system 1300, software application 1390, control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1450, 1480, flow diagram 2000, flow diagram 2100, flow diagram 2200, and / or computing device 2300.

[0271] For example, the aspects of therapy assessment 1430 shown in FIG. 15 and the aspects of the fourth dosing regimen 1460 shown in FIG. 18 can be similar. Similar reference numbers are used to indicate the features of the aspects of therapy assessment 1430 shown in FIG. 15 and the similar features of the aspects of the fourth dosing regimen 1460 shown in FIG. 18. One difference between the aspects of therapy assessment 1430 shown in FIG. 15 and the aspects of the fourth dosing regimen 1460 shown in FIG. 18 is that instead of having basal insulin, an unchanged phase 1436 with basal states 1436a - 1436h, and a maintenance titration phase 1438, the fourth dosing regimen 1460 includes a stepwise MDI of prandial insulin for multiple meals and a maintenance titration phase 1466 that includes a basal assessment 1467, a meal assessment 1468, and a correction ISF assessment 1469.

[0272] As shown in FIG. 18, the fourth dosing regimen 1460 can include a monitoring phase 1462, a rapid titration phase 1464, and a maintenance titration phase 1466. The monitoring phase 1462 can be configured to monitor one or more user compliance metrics of the current therapy over a predetermined period to ensure user adherence to the therapy. The rapid titration phase 1464 can be configured to introduce and apply a rapid dose of the fourth dosing regimen 1460 (e.g., a stepwise MDI of prandial insulin for multiple meals) and monitor the user's response to the additional therapy. The maintenance titration phase 1466 can be configured to apply the fourth dosing regimen 1460 (e.g., a stepwise MDI of prandial insulin for multiple meals) over a long period and monitor the user's response to the therapy.

[0273] As shown in FIG. 18, the maintenance titration phase 1466 can include a basal assessment 1467, a dietary assessment 1468, and / or a corrected ISF assessment 1469. In some embodiments, the basal assessment 1467, the dietary assessment 1468, and / or the corrected ISF assessment 1469 can be used with one or more dosing regimens (e.g., control plot 1400A) to determine a basal excess and / or to monitor any changes or side effects of the current therapy.

[0274] The basal assessment 1467 can be configured, for example, to monitor any changes to one or more basal states (e.g., basal states 1467a - 1467h shown in FIG. 19A) during the maintenance titration phase 1466. The basal assessment 1467 can be further configured to determine a basal excess of the current therapy and any changes or side effects. The basal assessment 1467 can be further configured to assist the software application 1390 in determining any trends over time (e.g., increase, decrease, no change) in the basal state and whether any dietary actions and / or corrective actions should be recommended.

[0275] The dietary assessment 1468 can be configured, for example, to monitor any changes to one or more dietary states (e.g., dietary states 1468a - 1468h shown in FIG. 19B) during the maintenance titration phase 1466. The dietary assessment 1468 can be further configured to determine a basal excess of the current therapy and any changes or side effects. The dietary assessment 1468 can be further configured to assist the software application 1390 in determining any trends over time (e.g., increase, decrease, no change) in the dietary state and whether any basal actions and / or corrective actions should be recommended.

[0276] The corrected ISF assessment 1469 can be configured, for example, to monitor any changes to one or more corrected ISF states (e.g., corrected ISF states 1469a-1469e shown in Figure 19C) during the maintenance titration phase 1466. The corrected ISF assessment 1469 can be further configured to determine basal excess of the current therapy and any changes or side effects. The corrected ISF assessment 1469 can be further configured to assist the software application 1390 in determining any trends over time in the corrected ISF state (e.g., increase, decrease, no change), and whether any basal actions and / or dietary actions should be recommended.

[0277] Figure 19A shows a basal assessment 1467 in an exemplary embodiment. Although the basal assessment 1467 is shown in Figure 19A as a standalone process, apparatus, and / or system, embodiments of the present disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14A–14C, 15–18, and 20–23, such as the therapy management system 1300, software application 1390, control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1450, 1460, 1480, flow diagram 2000, flow diagram 2100, flow diagram 2200, and / or computing device 2300.

[0278] As shown in Figure 19A, the basal assessment 1467 may include monitoring one or more basal states 1467a to 1467h. In some embodiments, one or more basal states 1467a to 1467h may include a first basal state 1467a (e.g., significantly below target), a second basal state 1467b (e.g., significantly below target), a third basal state 1467c (e.g., below target), a fourth basal state 1467d (e.g., at target), a fifth basal state 1467e (e.g., above target), a sixth basal state 1467f (e.g., significantly above target), a seventh basal state 1467g (e.g., significantly above target), an eighth basal state 1467h (e.g., basal excess), or a combination thereof.

[0279] In some embodiments, the first basal state 1467a can be determined based on a total glucose TBRm greater than approximately 4% or a glucose TVL of less than approximately 1%. In some embodiments, the second basal state 1467b can be determined based on an MMG of less than approximately 80 mg / dL or a treated nocturnal low of more than approximately 30 minutes. In some embodiments, the third basal state 1467c can be determined based on an MMG of approximately 80 mg / dL to approximately 100 mg / dL, or a BeAM of less than approximately -20 mg / dL / hr, or an MMG of approximately 100 mg / dL to approximately 130 mg / dL and a BBR of approximately 1.5 or higher. In some embodiments, the fourth basal state 1467d can be determined based on an MMG of approximately 100 mg / dL to approximately 130 mg / dL and a BBR of less than approximately 1.5. In some embodiments, the fifth basal state 1467e can be determined based on an MMG of approximately 130 mg / dL to approximately 150 mg / dL. In some embodiments, the sixth basal state 1467f can be determined based on an MMG of approximately 150 mg / dL to approximately 180 mg / dL. In some embodiments, the seventh basal state 1467g can be determined based on an MMG greater than approximately 180 mg / dL or a BBR of less than approximately 0.5. In some embodiments, the eighth basal state 1467h can be determined by a therapeutic assessment 1430 or by one or more of the above-mentioned basal excess metrics (e.g., TDD-to-body weight ratio, glucose variability, hypoglycemia metric, glucose change during meals, glucose change overnight).

[0280] Figure 19B shows a dietary assessment 1468 in an exemplary embodiment. Although the dietary assessment 1468 is shown in Figure 19B as a standalone process, apparatus, and / or system, embodiments of the present disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14A–14C, 15–18, and 20–23, such as the therapy management system 1300, software application 1390, control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1450, 1460, 1480, flow diagram 2000, flow diagram 2100, flow diagram 2200, and / or computing device 2300.

[0281] As shown in Figure 19B, the dietary assessment 1468 may include monitoring one or more dietary states 1468a to 1468h. In some embodiments, one or more dietary states 1468a to 1468h may include a first dietary state 1468a (e.g., significantly below target), a second dietary state 1468b (e.g., below target), a third dietary state 1468c (e.g., target), a fourth dietary state 1468d (e.g., above target), a fifth dietary state 1468e (e.g., significantly above target), a sixth dietary state 1468f (e.g., significantly above target), a seventh dietary state 1468g (e.g., saturated), an eighth dietary state 1468h (e.g., basal excess), or a combination thereof.

[0282] In some embodiments, the first dietary state 1468a can be determined based on a total glucose TRB of more than approximately 4%, a glucose TVL of less than approximately 1%, or an MPDG of less than approximately 70 mg / dL. In some embodiments, the second dietary state 1468b can be determined based on an MPDG of approximately 70 mg / dL to approximately 100 mg / dL, or a glucose TBR of more than approximately 4%. In some embodiments, the third dietary state 1468c can be determined based on an MPDG of approximately 100 mg / dL to approximately 120 mg / dL. In some embodiments, the fourth dietary state 1468d can be determined based on an MPDG of approximately 120 mg / dL to approximately 140 mg / dL. In some embodiments, the fifth dietary state 1468e can be determined based on an MPDG of approximately 140 mg / dL to approximately 180 mg / dL. In some embodiments, the sixth dietary state 1468f can be determined based on an MPDG of approximately 180 mg / dL to approximately 250 mg / dL. In some embodiments, the seventh dietary state 1468g can be determined based on an MPDG greater than approximately 250 mg / dL. In some embodiments, the eighth dietary state 1468h can be determined by the therapeutic assessment 1430 or by one or more of the basal excess metrics described above (e.g., TDD to body weight ratio, glucose fluctuation, hypoglycemia metric, glucose change during a meal, glucose change overnight).

[0283] Figure 19C shows an exemplary embodiment of the corrected ISF assessment 1469. Although the corrected ISF assessment 1469 is shown in Figure 19C as a standalone process, apparatus, and / or system, embodiments of the present disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14A–14C, 15–18, and 20–23, such as the therapy management system 1300, software application 1390, control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1450, 1460, 1480, flow diagram 2000, flow diagram 2100, flow diagram 2200, and / or computing device 2300.

[0284] As shown in Figure 19C, the corrected ISF assessment 1469 may include monitoring one or more corrected ISF states 1469a to 1469e. In some embodiments, one or more corrected ISF states 1469a to 1469e may include a first corrected ISF state 1469a (e.g., below target), a second corrected ISF state 1469b (e.g., at target), a third corrected ISF state 1469c (e.g., above target), a fourth corrected ISF state 1469d (e.g., significantly above target), a fifth corrected ISF state 1469e (e.g., basal excess), or a combination thereof.

[0285] In some embodiments, the first corrected ISF state 1469a can be determined based on a total glucose TBR greater than approximately 4%, a glucose TVL less than approximately 1%, or an MPDG less than approximately 70 mg / dL. In some embodiments, the second corrected ISF state 1469b can be determined based on a PRA-CG between approximately 70 mg / dL and approximately 180 mg / dL. In some embodiments, the third corrected ISF state 1469c can be determined based on a PRA-CG between approximately 180 mg / dL and approximately 250 mg / dL. In some embodiments, the fourth corrected ISF state 1469d can be determined based on a PRA-CG greater than approximately 250 mg / dL. In some embodiments, the fifth corrected ISF state 1469e can be determined by the therapy assessment 1430, or by one or more of the basal excess metrics described above (e.g., TDD-to-body weight ratio, glucose fluctuation, hypoglycemia metric, glucose change during meals, glucose change overnight).

[0286] Example flowchart Figure 20 shows a flowchart 2000 in an exemplary embodiment. For example, the flowchart 2000 may relate to a therapy management system 1300 shown in Figure 13. The flowchart 2000 can be configured to measure one or more metrics (e.g., glucose metric, therapy metric) of a first therapy (e.g., basal insulin) and to titrate the dose of the first therapy based at least in part on one or more metrics. The flowchart 2000 can be further configured to determine basal excess of the first therapy based on one or more metrics. If basal excess is determined, or if one or more user metrics remain outside one or more targets (e.g., one or more glucose metrics, A1C levels, body weight), the flowchart 2000 can be further configured to output a recommendation to add a second therapy (e.g., insulin analogue, GLP-1, prandial insulin, MDI of prandial insulin, etc.).

[0287] It should be understood that not all steps in Figure 20 are necessary to perform the disclosures provided herein. Furthermore, some steps may be performed simultaneously, sequentially, and / or in a different order than those shown in Figure 20. Flowchart 2000 will be illustrated with reference to Figures 13, 14A–14C, 15–18, 19A–19C, and 23. However, flowchart 2000 is not limited to its exemplary embodiments. Although the flow diagram 2000 is shown in Figure 20 as a standalone method, aspects of this disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, the elements in Figures 13, 14A-14C, 15-18, 19A-19C, and 21-23, such as control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1450, 1460, 1480, flow diagram 2100, flow diagram 2200, and / or computing device 2300. In some aspects, the flow diagram 2000 can be implemented by a software application 1390. In some aspects, the flow diagram 2000 can be implemented by one or more models or algorithms running on one or more processors and / or computing devices based on one or more instructions stored in one or more memories.

[0288] In step 2002, the user's glucose data can be received from an in vivo glucose monitoring device, as shown in Figures 13–18, 19A–19C, and 23. In some embodiments, glucose data may be received from the OBU 1350 of the therapy management system 1300 (e.g., via a software application 1390).

[0289] In step 2004, first therapy information can be received as shown in Figures 13–18, 19A–19C, and 23. In some embodiments, the first therapy may include basal insulin. In some embodiments, the first therapy information may be received from a drug delivery device of the therapy management system 1300 (e.g., a first injection pen 1360, a second injection pen 1370, an insulin pump, a smart pill bottle, etc.) (e.g., via a software application 1390). In some embodiments, the first therapy information may include one or more user metrics, including but not limited to user input, user feedback, data from drug delivery devices, body weight measurements, applied dose, TDD, TDD-to-body weight ratio, time between doses, type of drug, side effects of the first therapy, BMI, heart rate, blood pressure, diet data, exercise data, or a combination thereof.

[0290] In step 2006, one or more glucose metrics may be calculated based on glucose data, as shown in Figures 13-18, 19A-19C, and 23. In some embodiments, a software application 1390 of the therapy management system 1300 can receive glucose data and calculate one or more glucose metrics. In some embodiments, one or more glucose metrics may include mean glucose, glucose TIR, glucose TBR, glucose TVL, GMI, MMG, MPDG, PPG, BeAM, change in blood glucose during or overnight, postprandial glucose elevation for one or more meals, or a combination thereof.

[0291] In step 2008, the dose of basal insulin can be titrated based on one or more glucose metrics, as shown in Figures 13-18, 19A-19C, and 23. In some embodiments, the software application 1390 of the therapy management system 1300 can titrate the dose of basal insulin based on one or more glucose metrics.

[0292] In step 2010, as shown in Figures 13-18, 19A-19C, and 23, basal excess of the first therapy can be determined based on glucose data and one or more of the first therapy information. In some embodiments, the software application 1390 of the therapy management system 1300 can determine whether basal excess has occurred based on glucose data and one or more of the first therapy information. In some embodiments, the software application 1390 can determine basal excess based at least in part on the ratio of insulin TDD to the user's body weight, glucose fluctuations, hypoglycemia metrics, changes in blood glucose levels during meals or overnight, or a combination thereof.

[0293] In step 2012, if basal excess is determined, a recommendation to add a second therapy can be output, as shown in Figures 13-18, 19A-19C, and 23. In some embodiments, the software application 1390 of the therapy management system 1300 can output a recommendation to the user or HCP (e.g., via a remote device 1310) to add or initiate a second therapy. In some embodiments, the second therapy may include basal analogs, GLP-1 receptor agonists, dual GIP / GLP-1 receptor agonists, prandial insulin, MDI of prandial insulin, or a combination thereof.

[0294] In some embodiments, flow diagram 2000 can further include recommending reducing the basal insulin dose when a second therapy is added. In some embodiments, flow diagram 2000 can further include titrating the dose of the second therapy. In some embodiments, for example, titrating the dose of the second therapy can include receiving user input (e.g., feedback) regarding any side effects of the second therapy, and recommending an increase in the dose if the user input indicates no side effects over, for example, a monitoring period (e.g., 14 days). In some embodiments, for example, titrating the dose of the second therapy can include receiving the user's weight measurement (e.g., via a scale communicating with software application 1390), and recommending an increase in the dose if the weight measurement exceeds a target weight (e.g., 80 kg).

[0295] In some embodiments, flow diagram 2000 can further include recommending initiation of a prandial (bolus) insulin dose for the first meal if the post-meal glucose rise for one meal exceeds a threshold. In some embodiments, flow diagram 2000 can further include recommending initiation of a prandial (bolus) insulin dose (e.g., MDI) for multiple meals if the post-meal glucose rises for multiple meals exceed a threshold.

[0296] Exemplary non-adherence detection The user may intentionally or unintentionally fail to comply with the dosing guidance system and its recommendations. If the user is not taking the recommended dose, for example, skipping a dose or taking another dose, the titration algorithm is not effective. Further, if the system recommends a dose and the user's blood glucose remains high due to non-adherence, the system may continue to recommend increasingly higher doses. Then, if the user decides to take the dose after a period of non-compliance, the dose may be too high and not safe.

[0297] It would be beneficial to detect whether users are not adhering to dosage guidance and to encourage them to begin taking the recommended dose, or to provide guidance to educate users and improve therapy compliance.

[0298] Some embodiments described herein relate to determining non-adherence to therapeutic recommendations. A system may determine non-adherence based on an analysis of glucose metrics before and after a dose recommendation. A system may determine non-adherence based on determining glucose metrics before and after a dose recommendation, and determining whether there is a change in glucose metrics. A system may also determine adherence by determining expected changes and comparing the actual changes to the expected changes.

[0299] The system may determine the minimum glucose metric during a first period preceding the dose recommendation, or during a second period following the dose recommendation. The minimum glucose metric may be based on fasting blood glucose levels. The system may determine non-adherence if the change in the minimum glucose metric is less than the threshold change in the minimum glucose metric. This is because administration of the recommended dose is expected to result in a change in the minimum glucose metric. Therefore, no change or a minimal change may indicate that the user did not administer the recommended dose. The system may detect whether an increase in the dose recommendation resulted in a decrease in blood glucose or a decrease in the glucose metric. Alternatively, the system may detect whether a decrease in the dose recommendation resulted in an increase in blood glucose or an increase in the glucose metric. A change in the expected direction may be sufficient confirmation of adherence to the therapy, while an increase in blood glucose despite an increase in the dose recommendation may indicate non-adherence. The system may predict the expected range of blood glucose or glucose metric for a given dose recommendation. To determine adherence, the system may determine whether the actual blood glucose level after the dose recommendation is within the predicted range of values. When non-adherence is detected, the system may initiate one or more corrective actions.

[0300] The minimum glucose metric may be minimum morning glucose (MMG). The longest fasting period is typically the duration of a night's sleep. Therefore, MMG can be used as a substitute for fasting glucose by taking blood glucose levels in a morning window, such as 4 a.m. to 8 a.m., although other windows may be used. The use of MMG may require the system to have more information about the user, such as when the user is awake or asleep, to ensure that morning blood glucose is a good substitute for fasting blood glucose. Furthermore, if the user does not follow a typical schedule, such as a shift worker who works overnight, or if the user is traveling in a different time zone or maintains an irregular schedule, MMG may be an inadequate substitute for fasting glucose. Another metric is minimum hourly average glucose (DMHAG) of the day. Average blood glucose levels may be calculated hourly throughout the day, and the minimum hourly average of the day may be the lowest average blood glucose level of the day. This metric may be useful because it does not require the identification of a fasting period, or that the fasting period occurs at a specific time. However, DMHAG is applicable to patients receiving basal insulin-only therapy regimens because the rapid-acting insulin dose taken with meals, in addition to basal insulin, affects hypoglycemia.

[0301] The system described below can detect user non-adherence to dose recommendations based on glucose data. In some embodiments, the system can receive glucose data before and after administered doses to determine whether the user is adhering to the dose regimen. In some embodiments, the system can calculate one or more glucose metrics before and after administered doses. In some embodiments, the system can calculate one or more minimum glucose metrics before and after administered doses. In some embodiments, for example, the minimum glucose metric may be MMG. In some embodiments, for example, the minimum glucose metric may be DMHAG.

[0302] In some embodiments, the system can determine user adherence or non-adherence to a dose regimen (e.g., recommended dose) based on whether one or more minimum glucose metrics have changed by a predetermined amount (e.g., MMG outside the predicted level range). In some embodiments, the system can determine user adherence or non-adherence to a dose regimen (e.g., recommended dose) based on whether one or more glucose metrics have changed in the correct direction. For example, in the case of a dose increase, adherence can be determined by whether there is a measured decrease in blood glucose levels. In some embodiments, the system can determine user adherence or non-adherence by predicting what blood glucose levels and / or one or more glucose metrics should be after a dose has been administered, and whether the measured glucose data and / or one or more glucose metrics are outside the predicted range. In some embodiments, the system can determine user adherence or non-adherence based on a glucose data history for determining changes in blood glucose levels and / or changes in one or more glucose metrics for each administered dose. In some embodiments, for example, the system can predict and / or extrapolate the effects of a change in unit dose based on the determined change in blood glucose level for each administered dose, and / or the change in one or more glucose metrics for each administered dose.

[0303] Figure 21 shows a flowchart 2100 in an exemplary embodiment. For example, flowchart 2100 may relate to the therapy management system 1300 shown in Figure 13. Flowchart 2100 can be configured to measure one or more metrics (e.g., minimum glucose metric) of a first therapy (e.g., basal insulin) and to titrate the dose of the first therapy based at least in part on one or more metrics. Flowchart 2100 can be further configured to determine non-adherence by the user to one or more dose recommendations, e.g., non-adherence to one or more basal insulin dose recommendations. Flowchart 2100 can be further configured to output an indicator or notification (e.g., to the user, to a third party, etc.) in which basal insulin titration will be stopped if the change in the minimum glucose metric (e.g., fasting glucose, minimum daily glucose (DMG), minimum morning glucose (MMG), minimum hourly average daily glucose (DMHAG), etc.) is outside a predetermined metric.

[0304] It should be understood that not all steps in Figure 21 are necessary to perform the disclosures provided herein. Furthermore, some steps may be performed simultaneously, sequentially, and / or in a different order than those shown in Figure 21. Flowchart 2100 will be described with reference to Figures 13, 14A–14C, 15–18, 19A–19C, and 23. However, flowchart 2100 is not limited to its exemplary embodiments. Although the flow diagram 2100 is shown in Figure 21 as a standalone method, aspects of this disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14A-14C, 15-18, 19A-19C, 20, 22, and 23, such as control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1450, 1460, 1480, flow diagram 2000, flow diagram 2200, and / or computing device 2300. In some aspects, the flow diagram 2100 can be implemented by a software application 1390. In some aspects, the flow diagram 2100 can be implemented by one or more models or algorithms running on one or more processors and / or computing devices based on one or more instructions stored in one or more memories.

[0305] In step 2102, the user's glucose data can be received from an in vivo glucose monitoring device, as shown in Figures 13–18, 19A–19C, and 23. In some embodiments, glucose data can be received from the OBU 1390 of the therapy management system 1300 (for example, via a software application 1350).

[0306] In step 2104, the initial dose of basal insulin may be recommended, as shown in Figures 13-18, 19A-19C, and 23. In step 2106, one or more glucose metrics may be calculated based on glucose data, as shown in Figures 13-18, 19A-19C, and 23. One or more glucose metrics may include minimum glucose metrics. In some embodiments, a software application 1390 of the therapy management system 1300 can receive glucose data and calculate one or more glucose metrics. For example, one or more glucose metrics may be calculated for a period following a first dose recommendation. In some embodiments, one or more glucose metrics may include mean glucose, glucose TIR, glucose TBR, glucose TVL, GMI, MMG, MPDG, PPG, BeAM, change in blood glucose during or overnight, postprandial glucose elevation for one or more meals, or a combination thereof. In some embodiments, minimum glucose metrics may include fasting glucose, minimum daily glucose (DMG), minimum morning glucose (MMG), mean glucose over a given period, minimum hourly mean glucose (DMHAG), or a combination thereof.

[0307] In some embodiments, one or more minimum glucose metrics (e.g., fasting glucose, DMG, MMG, DMHAG, etc.) may be calculated based on glucose data. One or more minimum glucose metrics may be preferred over other glucose metrics because they provide a more accurate indicator of the user's blood glucose level. For example, MMG can be used as a substitute for fasting glucose and can typically be based on blood glucose levels in the morning window, which is the end of the longest fasting period of the night while the user is sleeping. Since MMG measures the user's fasting blood glucose level after the longest fasting period, it can provide a more accurate and safer indicator of the user's fasting blood glucose level.

[0308] In some embodiments, one or more minimum glucose metrics can be based on DMHAG. For example, DMHAG can be used as a substitute for fasting glucose and can be based on an hour of the day with the lowest average glucose level. DMHAG can be determined regardless of the fasting period or specific time of day and can provide a substitute indicator of a user's fasting blood glucose level. In some embodiments, DMHAG can be used as a safer substitute for MMG because it has similar accuracy to MMG but does not require meal and administration data. For example, DMHAG can be used for users with irregular schedules, such as shift workers who work overnight, travelers in various time zones, or any other users who maintain an irregular schedule.

[0309] In step 2108, the recommended dose of basal insulin can be titrated based on one or more glucose metrics, as shown in Figures 13-18, 19A-19C, and 23. In some embodiments, the software application 1390 of the therapy management system 1300 can titrate the recommended dose of basal insulin based on one or more glucose metrics. In some embodiments, the flow chart 2100 can titrate the recommended dose (e.g., via the software application 1390) based on one or more glucose metrics (e.g., via OBU 1350) and / or basal insulin dose-time data (e.g., via the first smart cap 1382 and / or the second smart cap 1384), without requiring basal insulin dose data.

[0310] In step 2110, a second dose of basal insulin may be recommended, as shown in Figures 13–18, 19A–19C, and 23. In some embodiments, the second dose may differ from the first dose, for example, due to titration of the recommended dose.

[0311] In step 2112, the change in the minimum glucose metric can be calculated as shown in Figures 13-18, 19A-19C, and 23. In some embodiments, the change in the minimum glucose metric can be calculated from a first period to a second period. In some embodiments, for example, the change can be calculated from a first period before recommending a second dose to a second period after recommending a second dose. In some embodiments, the software application 1390 of the therapy management system 1300 can receive glucose data and calculate the change in the minimum glucose metric. In some embodiments, the change in the minimum glucose metric can include changes in fasting glucose, DMG, MMG, average glucose over a given period, DMHAG, or a combination thereof.

[0312] In step 2114, it can be determined whether the change in the minimum glucose metric is outside a predetermined metric, as shown in Figures 13-18, 19A-19C, and 23. In some embodiments, the software application 1390 of the therapy management system 1300 can determine whether the change in the minimum glucose metric is outside a predetermined metric. In some embodiments, the predetermined metric may include the change in the minimum glucose metric being less than 5 mg / dL, less than 10 mg / dL, or less than 15 mg / dL, among other amounts. In some embodiments, the predetermined metric may be based on which minimum glucose metric is used. For example, the predetermined metric for MMG may include the change in MMG being less than 5 mg / dL, less than 10 mg / dL, or less than 15 mg / dL, among other amounts. In some embodiments, the predetermined metric for DMHAG may include the change in DMHAG being less than 0.1 mg / dL, less than 1 mg / dL, less than 5 mg / dL, or less than 10 mg / dL, among other amounts.

[0313] In some embodiments, a given metric may include the condition that the change in the minimum glucose metric is less than a predetermined percentage of the expected change in the minimum glucose metric. The predetermined percentage may be 50% of the expected change in the minimum glucose metric. In some embodiments, the expected change in the minimum glucose metric may be based on one or more user metrics. In some embodiments, one or more user metrics may include body weight, BMI, age, ISF, TDD, TDD-to-body weight ratio, A1C level, heart rate, blood pressure, or a combination thereof. Patients with higher insulin sensitivity will show smaller changes in blood glucose levels for a given dose of insulin compared to patients with lower insulin sensitivity. Furthermore, since body weight can be a surrogate for insulin sensitivity, increasing the dose by only 4U is expected to have less effect on blood glucose levels in patients with greater body weight than on the effect of a 4U dose administered to a person with relatively lower body weight. In some embodiments, the expected change in the minimum glucose metric may be based on the user's glucose data history. For example, if a previous dose increase of 1 U resulted in a 10 mg / dL decrease in blood glucose, the system may predict or anticipate that a further increase of 1 U will result in a further 10 mg / dL decrease in blood glucose. Furthermore, the system may determine or extrapolate trends based on historical data. For example, if a dose of 1 U previously resulted in a 10 mg / dL decrease in blood glucose, a dose of 2 U may be expected to result in a 20 mg / dL decrease. In some embodiments, the expected change in the minimum glucose metric may be proportional to the change between the first and second doses.

[0314] In some embodiments, a given metric may include a statistical test of the minimum glucose metric that does not support the hypothesis that the minimum glucose metric has changed. In some embodiments, the statistical test may include an insulin sensitivity test (IST), an insulin tolerance test (ITT), an oral glucose tolerance test (OGTT), a fasting plasma glucose (FPG) test, a random plasma glucose test, or a combination thereof.

[0315] In step 2116, as shown in Figures 13-18, 19A-19C, and 23, an indicator of basal insulin titration with cessation can be output if the change in the minimum glucose metric is outside a predetermined metric. In some embodiments, the software application 1390 of the therapy management system 1300 can output an indicator to the user or HCP (e.g., via a remote device 1310) that basal insulin titration will be stopped (e.g., to avoid dangerous administration events, hypoglycemic events, hyperglycemic events).

[0316] In some embodiments, the flowchart 2100 may output a notification to the user. The notification may indicate that the user has not taken the recommended dose. The notification may encourage the user to administer the recommended dose. The notification may prompt the user to confirm the administration of a second dose. In some embodiments, the flowchart 2100 may further include resuming dose guidance to the user once the user has confirmed the administration of the second dose. In some embodiments, the flowchart 2100 may further include recommending a third dose based on the titration of the second dose if the user has not confirmed the administration of the second dose. The notification may provide educational materials or guide the user to educational materials regarding the user's therapy and / or the operation of the dose guidance system.

[0317] In some embodiments, the flowchart 900 may further include outputting prompts to the user to confirm that the user is following the recommended dosage. In some embodiments, the flowchart 900 may further include outputting prompts to the user to prompt the user to seek guidance from a healthcare professional. In some embodiments, the flowchart 900 may further include outputting a quiz to the user to verify that the user is following the recommended dosage. In some embodiments, for example, a software application 190 of the therapy management system 100 may output a quiz to the user (for example, via a remote device 110) (e.g., "What is your current basal insulin dose?", "What was your last basal insulin dose?", "What is the recommended basal insulin dose?"). If the user answers correctly, titration may continue. If the user enters a dosage different from the recommended dosage, the titration algorithm may proceed based on the dosage entered by the user as the dosage for titration.

[0318] In some embodiments, flow chart 2100 may further include recommending limiting the change in basal insulin dose if there is no change in the minimum glucose metric after titrating the recommended dose over a predetermined period. In some embodiments, the predetermined period may include at least 3 days. In some embodiments, the predetermined period may include at least 5 days. In some embodiments, the predetermined period may include at least 7 days. In some embodiments, the predetermined period may range from 1 to 14 days.

[0319] In some embodiments, the flow chart 2100 may further include recommending that upward titration (i.e., increasing the dose) of basal insulin be stopped if the minimum glucose metric (e.g., fasting glucose, MMG, DMHAG, etc.) does not decrease after titrating the recommended dose over a predetermined period. In some embodiments, the flow chart 2100 may further include recommending that downward titration (i.e., decreasing the dose) of basal insulin be stopped if the minimum glucose metric (e.g., fasting glucose, MMG, DMHAG, etc.) does not increase after titrating the recommended dose over a predetermined period.

[0320] In some embodiments, the flow chart 2100 may further include activating a blind mode so that glucose data is not displayed to the user. In some embodiments, the blind mode can prevent the display of glucose data to the user, thereby encouraging the user to adhere to the dose regimen, reducing inappropriate administration and / or user interaction, and correcting the user's blood glucose levels (e.g., preventing the user from deviating from the recommended dose).

[0321] In some embodiments, the flow chart 2100 may further include initiating a counter configured to monitor the number of titration cycles if the change in the minimum glucose metric is outside a predetermined metric. In some embodiments, the flow chart 2100 may further include recommending cessation of basal insulin titration if, after a predetermined number of titration cycles in which the dose of basal insulin has been increased or decreased, the change in the minimum glucose metric remains outside the predetermined metric.

[0322] In some embodiments, the flow chart 2100 may further include receiving basal insulin dose-time data. In some embodiments, basal insulin dose-time data may be determined based on changes in blood glucose levels (e.g., received from OBU 1350) via the software application 1390 and / or received from the drug delivery device and / or pen caps (e.g., first smart cap 1382, second smart cap 1384) of the therapy management system 1300 (e.g., via the software application 1390). In some embodiments, basal insulin dose-time data may supplement one or more glucose metrics when determining the recommended dose.

[0323] In some embodiments, the flow chart 2100 can detect non-adherence to the recommended basal insulin dose. In some embodiments, the flow chart 2100 can prevent unsafe continuous increases or decreases in the recommended dose that could lead to an unsafe basal insulin dose. In some embodiments, the flow chart 2100 may include one or more outputs (e.g., indicators, notifications, prompts, etc.) for seeking guidance from the user's HCP to confirm that the user is following the dose recommendation, and / or other verification means (e.g., quizzes, queries, etc.) to confirm that the user is following the dose recommendation.

[0324] In some embodiments, non-adherence assessment may be applied to rapid-acting doses for meals. However, the glucose metric will likely be based on the glucose metric in the post-prandial period following the meal dose, rather than on MMG or DMHAG. To determine changes in the metric, the glucose metric in the post-prandial period for meals may be compared before and after the new dose recommendation.

[0325] Figure 22 shows a flowchart 2200 in an exemplary embodiment. For example, flowchart 2200 may relate to the therapy management system 1300 shown in Figure 13. Flowchart 2200 can be configured to measure one or more metrics (e.g., minimum glucose metric) of a therapy (e.g., basal insulin) to determine non-adherence by the user to one or more dose recommendations. Flowchart 2200 can be further configured to output an indicator or notification of non-adherence by the user to a third party (e.g., the user, a third party, etc.).

[0326] It should be understood that not all steps in Figure 22 are necessary to perform the disclosures provided herein. Furthermore, some steps may be performed simultaneously, sequentially, and / or in a different order than those shown in Figure 22. Flowchart 2200 will be described with reference to Figures 13, 14A–14C, 15–18, 19A–19C, and 23. However, flowchart 2200 is not limited to its exemplary embodiments. Although the flow diagram 2200 is shown in Figure 22 as a standalone method, aspects of this disclosure can be used in conjunction with other processes, apparatus, systems, and / or methods, for example, but not limited to, elements in Figures 13, 14A-14C, 15-18, 19A-19C, 20, 21, and 23, such as control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1450, 1460, 1480, flow diagram 2000, flow diagram 2100, and / or computing device 2300. In some aspects, the flow diagram 2200 can be implemented by a software application 1390. In some aspects, the flow diagram 2200 can be implemented by one or more models or algorithms running on one or more processors and / or computing devices based on one or more instructions stored in one or more memories.

[0327] In step 2202, the user's glucose data for the first period may be received from an in vivo glucose monitoring device, as shown in Figures 13–18, 19A–19C, and 23. In some embodiments, glucose data may be received from the OBU 1390 of the therapy management system 1300 (e.g., via the software application 1350). In some embodiments, the first period may include a morning window or a fasting period.

[0328] In step 2204, a first minimum glucose metric can be determined for a first period, as shown in Figures 13-18, 19A-19C, and 23. In some embodiments, a software application 1390 of the therapy management system 1300 can receive glucose data for the first period and calculate the first minimum glucose metric. In some embodiments, the first minimum glucose metric may include fasting glucose, DMG, MMG, average glucose over a given period, DMHAG, or a combination thereof.

[0329] In step 2206, dose recommendations can be provided to the user based on glucose data during a first period, as shown in Figures 13–18, 19A–19C, and 23. In some embodiments, the software application 1390 of the therapy management system 1300 can calculate dose recommendations based on one or more glucose metrics (e.g., minimum glucose metric) of blood glucose levels and / or glucose data during a first period.

[0330] In step 2208, as shown in Figures 13–18, 19A–19C, and 23, the user's glucose data for a second period after dose recommendation may be received from an in vivo glucose monitoring device. In some embodiments, glucose data can be received from the OBU 1390 of the therapy management system 1300 (e.g., via the software application 1350). In some embodiments, the second period may include a window after dose recommendation has been provided. In some embodiments, the second period may include a window after the user has confirmed administration of the recommended dose.

[0331] In step 2210, a second minimum glucose metric can be determined for a second period, as shown in Figures 13–18, 19A–19C, and 23. In some embodiments, a software application 1390 of the therapy management system 1300 can receive glucose data for the second period and calculate the second minimum glucose metric. In some embodiments, the second minimum glucose metric may include fasting glucose, DMG, MMG, average glucose over a given period, DMHAG, or a combination thereof. In some embodiments, the first and second minimum glucose metrics may be of the same type (e.g., both are MMG, both are DMHAG, etc.). In some embodiments, the first and second minimum glucose metrics may be different (e.g., the first minimum glucose may be MMG and the second minimum glucose may be DMHAG, etc.).

[0332] In step 2212, non-adherence to a dose recommendation can be determined based on a comparison of the first and second minimum glucose metrics, as shown in Figures 13–18, 19A–19C, and 23. In some embodiments, the comparison may be based on whether the first and second minimum glucose metrics change by a predetermined amount (e.g., the change in MMG is outside the range of expected change). In some embodiments, the comparison may be based on whether the first and second minimum glucose metrics change in the correct direction. For example, in the case of a dose recommendation to increase the dose, non-adherence can be determined if no subsequent decrease in blood glucose is measured. In some embodiments, the software application 1390 of the therapy management system 1300 can receive glucose data for the first and second periods, determine the first and second minimum glucose metrics, and compare the first and second minimum glucose metrics.

[0333] In step 2214, as shown in Figures 13-18, 19A-19C, and 23, an indicator of non-adherence can be output, for example, based on comparison. In some embodiments, the software application 1390 of the therapy management system 1300 can output the indicator of non-adherence to the user and / or HCP (e.g., a third party) (for example, via a remote device 1310).

[0334] In some embodiments, the software application 1390 may include one or more models or algorithms (e.g., those described herein) that run on one or more processors and / or computing devices based on one or more instructions stored in one or more memories. In some embodiments, the software application 1390 may be stored in and / or run on a telephone (e.g., remote device 1310), a drug delivery device (e.g., a first injection pen 1360, a second injection pen 1370), a pen cap (e.g., a first smart cap 1382, a second smart cap 1384), a cloud server, or a combination thereof. In some embodiments, one or more of the algorithms described above for titration and / or dose recommendation (e.g., control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, flow diagram 2000, flow diagram 2100, etc.) can be stored and / or executed on a telephone (e.g., remote device 1310), a drug delivery device (e.g., first injection pen 1360, second injection pen 1370), a pen cap (e.g., first smart cap 1382, second smart cap 1384), a cloud server, or a combination thereof.

[0335] In some embodiments, one or more therapy regimens may include premixed insulin, which may include a combination of long-acting (LA) insulin and rapid-acting (RA) insulin. In some embodiments, basal insulin therapy and / or prandial insulin therapy may utilize premixed insulin. For example, a third dosing regimen 1450 may utilize premixed insulin for a single meal for a single combined dose of LA (basal) insulin and RA (bolus) insulin.

[0336] In some embodiments, the software application 1390 may include one or more titration algorithms. In some embodiments, the software application 1390 may utilize multiple titration algorithms (e.g., general, personalized, optimized). In some embodiments, the software application 1390 may provide the user with options for selecting a particular titration algorithm. In some embodiments, the software application 1390 may include general titration algorithms based on American Diabetes Association (ADA) guidelines (e.g., fasting blood glucose levels of 80–130 mg / dL, starting insulin dose of 10 units / day or 0.1–0.2 units / kg / day, insulin dose increments of 5–15% or 1–4 units, etc.). In some embodiments, the software application 1390 may include a personalized titration algorithm tailored to the user based on one or more glucose metrics and / or one or more user metrics, such as those described herein (e.g., control diagram 1400A, therapy assessment 1400B, control diagram 1400C, therapy assessment 1430, flow diagram 2000, flow diagram 2100, etc.). In some embodiments, the software application 1390 may include an optimized titration algorithm based on AI and / or machine learning, which may generate a constructed user profile using, for example, an AI module trained using analytical methods. In some embodiments, the optimized titration algorithm can utilize simulated annealing, gradient descent, finite differences, interpolation, population models, regression, parameter adaptation, supervised machine learning, unsupervised machine learning, neural networks, classification models, clustering, vector quantization, stochastic gradient descent, implicit updates, leak averaging, momentum methods, adaptive gradients (AdaGrad), backpropagation, root mean square propagation (RMSProp), adaptive moment estimation (Adam), or a combination thereof.

[0337] Exemplary computing device Figure 23 shows the computing device 2300 in various exemplary embodiments. The computing device 2300 can be configured to implement the operations described herein. The computing device 2300 can be further configured to run the operating systems, application programs, control diagrams, flow diagrams, and / or software modules (e.g., software engines) described herein. For example, the computing device 2300 can be configured to implement the control diagram 1400A shown in Figure 14A. For example, the computing device 2300 can be configured to implement the flow diagram 2000 shown in Figure 20. For example, the computing device 2300 can be configured to implement the flow diagram 2100 shown in Figure 21. For example, the computing device 2300 can be configured to implement the flow diagram 2200 shown in Figure 22. Although the computing device 2300 is shown in Figure 23 as a standalone device and / or system, aspects of the present disclosure can be used with other devices, systems, and / or methods, for example, but not limited to, Figures 14A to 18, Figures 19A to 19C, and Figures 20 to 22, such as control diagram 1400A, therapy assessment 1430, one or more dosing regimens 1410, 1440, 1450, 1460, 1480, flow diagram 2000, flow diagram 2100, and / or flow diagram 2200.

[0338] For example, an embodiment of computing device 1200 shown in Figure 12 and an embodiment of computing device 2300 shown in Figure 23 may be similar. Similar reference numerals are used to indicate the features of the embodiment of computing device 1200 shown in Figure 12 and similar features of the embodiment of computing device 2300 shown in Figure 23. Computing device 1200 can implement the operations described herein and run the operating systems, application programs, control diagrams, flowcharts, and / or software modules (e.g., software engines) described herein.

[0339] As shown in Figure 23, the computing device 2300 may include a processing device 2302 (e.g., one or more processors, one or more microprocessors, microcontrollers), system memory 2304 (e.g., one or more memories), system bus 2306, read-only memory (ROM) 2308, random access memory (RAM) 2310, basic input / output system (BIOS) 2312, secondary storage device 2314 (e.g., hard drive), secondary storage interface 2316, operating system (OS) 2318, application program (app) 2320, program module 2322 (e.g., software engine, algorithm), program data 2324, input device 2326 (e.g., keyboard 2328, mouse 2330, microphone 2332, touch sensor 2334, gesture sensor 2335), input / output (I / O) interface 2336, display device 2338, video adapter 2340, and network interface 2342.

[0340] In some embodiments, the computing device 2300 may include at least one processing device 2302 (e.g., a processor), such as a central processing unit (CPU). Various processing devices are available from various manufacturers, e.g., Intel or Advanced Micro Devices. In this example, the computing device 2300 also includes system memory 2304 and a system bus 2306 that connects various system components, including the system memory 2304, to the processing device 2302. The system bus 2306 may be any number of bus structures that can be used, including but not limited to a memory bus or memory controller, a peripheral bus, or a local bus using any of various bus architectures.

[0341] Examples of computing devices that can be implemented using computing device 2300 include desktop computers, laptop computers, tablet computers, mobile computing devices (such as smartphones, touchpad mobile digital devices, or other mobile devices), or other devices configured to process digital instructions.

[0342] The system memory 2304 includes ROM 2308 and RAM 2310. For example, during startup, the BIOS 2312, which contains basic routines that function to transfer information within the computing device 2300, may be stored in ROM 2308.

[0343] In some embodiments, the computing device 2300 may include a secondary storage device 2314, such as a hard disk drive, for storing digital data. The secondary storage device 2314 may be connected to the system bus 2306 by a secondary storage interface 2316. The secondary storage device 2314 and its associated computer-readable medium can provide a non-volatile and non-temporary storage device for computer-readable instructions, data structures, and other data for the computing device 2300 (e.g., application programs and program modules).

[0344] The exemplary environments described herein employ a hard disk drive as the secondary storage device, but other embodiments may use other types of computer-readable storage media. Examples of these other types of computer-readable storage media include magnetic cassettes, flash memory cards, digital video discs, Bernoulli cartridges, compact disc read-only memory, digital multipurpose disc read-only memory, random access memory, or read-only memory. Some embodiments may include non-temporary media. For example, computer program products can be tangibly embodied in non-temporary storage media. In addition, such computer-readable storage media may include local storage devices or cloud-based storage devices.

[0345] Several program modules, including OS 2318, one or more application programs 2320, other program modules 2322 (such as software applications and software engines as described herein), and program data 2324, can be stored in the secondary storage device 2314 and / or system memory 2304. The computing device 2300 may utilize any suitable operating system, such as Microsoft Windows®, Google Chrome® OS, Apple OS, Unix or Linux and its variations, and any other operating system suitable for the computing device. Other examples include operating systems from Microsoft, Google, or Apple, or any other suitable operating systems used in tablet computing devices.

[0346] In some embodiments, a user can provide one or more inputs to the computing device 2300 via one or more input devices 2326. Examples of input devices 2326 include a keyboard 2328, a mouse 2330, a microphone 2332 (e.g., for voice and / or other audio input), a touch sensor 2334 (e.g., a touchpad and / or touch-sensitive display), and a gesture sensor 2335 (e.g., for gesture input). In some embodiments, the input devices 2326 can provide detection based on presence, proximity, and / or movement. In some embodiments, a user can walk into the house, which can trigger an input to the processing device 2302. For example, the input device 2326 then facilitates an automated experience for the user. Other embodiments may include other input devices 2326. The input devices 2326 can be connected to the processing device 2302 via an I / O interface 2336 that can be coupled to the system bus 2306. The input device 2326 can be connected by any number of I / O interfaces, such as a parallel port, serial port, game port, and / or a universal serial bus. In some embodiments, the input device 2326 can wirelessly communicate with I / O interfaces 2336, including, for example, infrared, Bluetooth® wireless technology, 802.11a / b / g / n, cellular, ultra-wideband (UWB), ZigBee, or other radio frequency (RF) communication systems.

[0347] In this exemplary embodiment, a display device 2338, such as a monitor, liquid crystal display device, light-emitting diode display device, projector, or touch-sensitive display device, can also be connected to the system bus 2306 via an interface such as a video adapter 2340. In some embodiments, the computing device 2300 may include various other peripheral devices (not shown), such as speakers or printers, in addition to the display device 2338.

[0348] The computing device 2300 can connect to one or more networks via a network interface 2342. The network interface 2342 can provide wired and / or wireless communication. In some embodiments, the network interface 2342 may include one or more antennas for transmitting and / or receiving wireless signals. In some embodiments, when used in a local area networking environment or a wide area networking environment (e.g., the Internet), the network interface 2342 may include an Ethernet interface. Other possible embodiments may use other communication devices. For example, some embodiments of the computing device 2300 may include a modem for communication across the entire network.

[0349] The computing device 2300 may include at least some form of computer-readable medium. The computer-readable medium may include any available medium accessible by the computing device 2300. For example, the computer-readable medium may include computer-readable storage medium and computer-readable communication medium.

[0350] Computer-readable storage media may include volatile and non-volatile, removable and non-removable media configured to store information such as computer-readable instructions, data structures, program modules, or other data, and implemented within any device. Examples of computer-readable storage media include, but are not limited to, random-access memory, read-only memory, electrically erasable and writable read-only memory, flash memory or other memory technologies, compact disk read-only memory, digital multipurpose disk or other optical storage devices, magnetic cassettes, magnetic tapes, magnetic disk storage devices or other magnetic storage devices, or any other media that can be used to store desired information and are accessible by the computing device 2300.

[0351] Computer-readable communication media may include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information transmission medium. The term “modulated data signal” refers to a signal in which one or more of its characteristics are set or modified in a manner that encodes the information in the signal. For example, computer-readable communication media may include wired media such as wired networks or direct wired connections, as well as wireless media such as acoustic, radio frequency, infrared, and other wireless media. Any combination of the above is also included in the scope of computer-readable media.

[0352] The computing device 2300 shown in Figure 23 is also an example of a programmable electronic device that may include one or more such computing devices, and if multiple computing devices are included, such computing devices may be coupled together into a suitable data communication network to collectively perform various functions, methods, or operations disclosed herein.

[0353] Various embodiments of this disclosure can be implemented on a computing device by software, firmware, hardware, or a combination thereof. Figure 23 shows an exemplary computing device 2300 in which an intended embodiment or part thereof can be implemented as computer-readable code. For example, the method shown by the control diagram described herein can be implemented in computing device 2300. Various embodiments are described in terms of this exemplary computing device 2300. Those skilled in the art will see from the above description how these embodiments can be implemented using other computer systems and / or computer architectures.

[0354] The expressions and terms used herein are for illustrative purposes only, not limitation, and should be understood to be interpreted by those skilled in the art in light of the teachings herein.

[0355] The above examples illustrate, but do not limit, aspects of the present disclosure. Various other suitable modifications and adaptations of conditions and parameters, which are commonly encountered in the art and would be apparent to those skilled in the art, are within the intent and scope of this disclosure.

[0356] While specific embodiments have been described above, it should be understood that these embodiments may be implemented in forms other than those described. The description is not intended to limit the scope of the claims.

[0357] It should be understood that the Detailed Description section, rather than the Summary and Abstract section, is intended to be used to interpret the claims. The Summary and Abstract section may describe one or more, but not all, exemplary embodiments as contemplated by the inventors(s), and is therefore not intended to limit the embodiments and appended claims in any way.

[0358] The embodiments have been described above using functional building blocks that represent implementation forms of specific functions and their relationships. The boundaries of these functional building blocks are arbitrarily defined in this specification for the sake of explanation. Alternative boundaries may be defined, as long as the specific functions and their relationships are properly performed.

[0359] While the foregoing descriptions of specific embodiments will adequately illustrate the general nature of those embodiments, the general nature of those embodiments can be readily modified and / or adapted for various uses of such specific embodiments by others applying knowledge within the scope of the art, without excessive experimentation and without deviating from the general concept of these embodiments. Therefore, such adapted and modified forms are intended to be within the meaning and scope of equivalents of the disclosed embodiments, based on the teachings and guidance presented ...

Claims

1. A non-temporary computer-readable storage medium that stores instructions causing the processor to perform operations for an insulin therapy management process when executed by the processor, wherein the operations are: A process of conducting a survey of diabetic patients (PWD) and receiving survey information in accordance with the said survey, A step of performing a high-touch dialogue phase with respect to the PWD, wherein the high-touch dialogue phase includes iterative adjustment of the insulin therapy based on collected data and published guidelines for the insulin therapy, and the high-touch dialogue phase establishes at least a basal dose setting, a meal dose setting, and a corrected dose setting for the PWD. A non-temporary computer-readable storage medium comprising: a step of performing a progress phase relating to the PWD after the completion of the high-touch interaction phase, wherein the progress phase includes a step of collecting performance metrics relating to the PWD, a step of applying reinforcement learning to the performance metrics, and a step of automatically performing actions relating to the PWD based on the reinforcement learning.

2. The non-temporary computer-readable storage medium according to claim 1, wherein the high-touch interaction phase is performed over a predetermined period of time.

3. The non-temporary computer-readable storage medium according to claim 1, wherein the high-touch interaction phase has a target-based duration.

4. The non-temporary computer-readable storage medium according to claim 3, wherein the target-based duration depends on the determination that the blood glucose level of the PWD is within the target range and that the PWD is being administered insulin as specified in the insulin therapy.

5. The non-temporary computer-readable storage medium according to claim 1, wherein the operation further comprises the steps of evaluating each of a plurality of insights and determining whether the insight has been triggered, taking into account at least the performance metrics and the collected data, and the action is selected on the basis that one or more of the insights have been triggered.

6. The non-temporary computer-readable storage medium according to claim 5, wherein the insight is evaluated in accordance with its priority over the PWD.

7. The non-temporary computer-readable information according to claim 6, wherein the insights to be evaluated are selected from a set of insights based on having the highest priority.

8. The non-temporary computer-readable storage medium according to claim 5, wherein the step of evaluating one or more of the aforementioned insights includes a multimodal evaluation.

9. The non-temporary computer-readable storage medium according to claim 1, wherein the operation further includes the step of performing an observation regarding the state of the PWD after automatically performing the action.

10. The non-temporary computer-readable storage medium according to claim 9, wherein the reinforcement learning includes the step of providing positive or negative feedback for the selection of the action.

11. The non-temporary computer-readable storage medium according to claim 10, wherein the operation further comprises the step of performing a cool-down after the completion of the observation.

12. The non-temporary computer-readable storage medium according to claim 11, wherein the operation further comprises the steps of evaluating each of a plurality of insights and determining whether the insight has been triggered, taking into account at least the performance metric and the collected data, and the action is selected on the basis that one or more of the insights have been triggered.

13. The non-temporary computer-readable storage medium according to claim 12, further comprising the step of checking whether the cooldown is accompanied by the action before selecting the action.

14. If the cooldown is accompanied by the action, the operation instead evaluates the next of the plurality of insights, according to claim 13, a non-temporary computer-readable storage medium.

15. The non-temporary computer-readable storage medium according to claim 9, wherein the operation further comprises the steps of evaluating each of a plurality of insights and determining whether the insight has been triggered, taking into account at least the performance metric and the collected data, and the action is selected on the basis that one or more of the insights have been triggered.

16. The non-temporary computer-readable storage medium according to claim 15, further comprising the step of checking whether the observation is accompanied by the insight before selecting the action.

17. If the observation is accompanied by the insight, the operation evaluates improvement criteria with respect to the insight, taking into account at least the performance metrics and the collected data, the non-temporary computer-readable storage medium according to claim 14.

18. The non-temporary computer-readable storage medium according to claim 17, wherein the evaluation of the aforementioned improvement criteria includes multimodal evaluation.

19. The non-temporary computer-readable storage medium according to claim 14, wherein for any of the insights that have not been triggered in the evaluation, the operation further comprises the steps of determining whether the insight has not been triggered for at least a predetermined period of time, and performing a streak action taking into account the determination.

20. The non-temporary computer-readable storage medium according to claim 1, wherein the step of automatically performing the aforementioned action includes the step of selecting one or more targets for the PWD and the step of presenting the one or more targets to the PWD.

21. The non-temporary computer-readable storage medium according to claim 20, wherein the operation further includes the step of receiving an opt-in input or an opt-out input by the PWD.

22. A method for escalating therapy for diabetic patients, wherein the method is: The process involves receiving user glucose data from an in vivo glucose monitoring device, A step of receiving first therapeutic information for a first therapy, wherein the first therapy includes basal insulin, A step of calculating one or more glucose metrics based on the received glucose data, A step of titrating the dose of basal insulin based on the one or more glucose metrics, A method comprising the step of determining basal excess based on the glucose data and one or more of the first therapeutic information.

23. The method according to claim 22, further comprising the step of outputting a recommendation to add a second therapy if basal excess is detected.

24. The method according to claim 22, wherein the one or more glucose metrics include mean glucose, median glucose, glucose time-in-range (TIR), glucose time-below-range (TBR), glucose time-above-range (TAR), glucose time-very-low (TVL), glucose management index (GMI), minimum wake glucose (MMG), minimum post-administration glucose (MPDG), postprandial glucose (PPG), sleep-to-wake glucose (BeAM), or a combination thereof.

25. The method according to claim 22, wherein the first therapeutic information is manually entered via input from a computing device.

26. The method according to claim 22, wherein the first therapeutic information is collected by a drug delivery device and communicated to a computing device.

27. The method according to claim 22, wherein basal excess is determined based on the ratio of the total daily dose (TDD) of insulin to the user's body weight.

28. The method according to claim 22, wherein basal excess is determined based on glucose variability.

29. The method according to claim 22, wherein basal excess is determined based on a hypoglycemic metric.

30. The method according to claim 22, wherein basal excess is determined based on changes in blood glucose levels during a meal or overnight.

31. The method of claim 22, further comprising the step of reducing the dose of basal insulin if a second therapy is added.

32. The method according to claim 22, wherein the second therapy comprises a glucagon-like peptide-1 (GLP-1) receptor agonist.

33. The process further comprises a step of titrating the dose of the GLP-1 receptor agonist, and the step of titrating the dose is: A step of receiving user input regarding the side effects of the second therapy, The method according to claim 32, comprising the step of recommending an increase in dose if the user input is associated without side effects.

34. The process further comprises a step of titrating the dose of the GLP-1 receptor agonist, and the step of titrating the dose is: The process of receiving the user's weight measurement, The method according to claim 32, further comprising the step of recommending an increase in dosage if the weight measurement exceeds the target weight.

35. The method according to claim 22, wherein the one or more glucose metrics include a post-meal glucose increase for each of a plurality of meals.

36. The method according to claim 35, further comprising the step of recommending the initiation of a pranial insulin dose for the first meal if the post-meal glucose elevation for a single meal exceeds a threshold.

37. The method according to claim 35, further comprising the step of recommending the initiation of a pranial insulin dose for each of the multiple meals if the post-meal glucose elevation exceeds a threshold for each of the multiple meals.

38. A system for escalating therapies for diabetic patients, wherein the system is An in vivo glucose monitoring device configured to measure the user's glucose data, A remote device that communicates with the in vivo glucose monitoring device, and is configured to receive or retrieve glucose data from the in vivo glucose monitoring device, A drug delivery device that communicates with the in vivo glucose monitoring device and the remote device, the drug delivery device being configured to administer one or more administration regimens, A processor that communicates with a sample measurement system, the remote device, and the drug delivery device, wherein the processor is coupled to a memory for storing instructions, and when an instruction is executed, the processor receives The process of receiving the user's glucose data from the in vivo glucose monitoring device, A step of receiving first therapeutic information of a first therapy from the remote device, the drug delivery device, or both, wherein the first therapy includes basal insulin; A step of calculating one or more glucose metrics based on the received glucose data, A step of titrating the dose of basal insulin based on the one or more glucose metrics, A system comprising a processor that performs an operation including the step of determining basal excess based on the glucose data and one or more of the first therapeutic information.

39. The system according to claim 38, further comprising the step of outputting a recommendation to the remote device to add a second therapy if basal excess is detected.

40. The system according to claim 38, wherein the one or more glucose metrics include mean glucose, median glucose, glucose TIR, glucose TBR, glucose TAR, glucose TVL, GMI, MMG, MPDG, PPG, BeAM, post-meal glucose elevation for each of a plurality of meals, or a combination thereof.

41. The system according to claim 38, wherein basal excess is determined based on the ratio of insulin TDD to the user's body weight.

42. The system according to claim 38, wherein basal excess is determined based on glucose fluctuations, hypoglycemia metrics, changes in blood glucose levels during a meal or overnight, or a combination thereof.

43. The second therapy is the system according to claim 38, comprising a GLP-1 receptor agonist, a prandial insulin administration with the first meal, prandial insulin doses with multiple meals, or a combination thereof.

44. A method for escalating therapy for diabetic patients to detect non-adherence to basal insulin recommendations, wherein the method is: The process involves receiving user glucose data from an in vivo glucose monitoring device, A step of recommending a first dose of basal insulin to the user, A step of calculating one or more glucose metrics based on the received glucose data, wherein the one or more glucose metrics include a minimum glucose metric, A step of titrating the recommended dose of basal insulin based on the one or more glucose metrics, A step of recommending a second dose of basal insulin to the user, wherein the second dose differs from the first dose due to titration of the recommended dose, A step of calculating the change in the minimum glucose metric from a first period before recommending the second dose to a second period after recommending the second dose, A step of determining whether the change in the minimum glucose metric is outside a predetermined metric, A method comprising the step of outputting an indicator that titration of basal insulin has stopped if the change in the minimum glucose metric is outside the predetermined metric.

45. The method according to claim 44, wherein the minimum glucose metric includes the minimum hourly average glucose (DMHAG) of the day.

46. The method according to claim 44, wherein the predetermined metric includes the change in the minimum glucose metric being less than a predetermined percentage of the expected change in the minimum glucose metric.

47. The method according to claim 46, wherein the expected change in the minimum glucose metric is based on one or more user metrics.

48. The method according to claim 47, wherein the one or more user metrics include an insulin sensitivity factor (ISF).

49. The method according to claim 46, wherein the expected change in the minimum glucose metric is based on the user's glucose data history.

50. The method according to claim 46, wherein the expected change in the minimum glucose metric is proportional to the change between the first dose and the second dose.

51. The method according to claim 44, wherein the predetermined metric includes the change in the minimum glucose metric being less than a predetermined change in blood glucose level.

52. The method according to claim 44, wherein the predetermined metric is determined based on a statistical method of the change in the minimum glucose metric.

53. The method according to claim 44, further comprising the step of prompting the user to confirm the administration of the second dose.

54. The method according to claim 53, further comprising the step of restarting titration when the user confirms the administration of the second dose.

55. The method according to claim 53, further comprising the step of recommending a third dose based on the titration of the second dose if the user has not confirmed the administration of the second dose.

56. The method according to claim 44, further comprising the step of recommending that if there is no change in the minimum glucose metric after titrating the recommended dose over a predetermined period, the change in the basal insulin dose be limited.

57. The method according to claim 56, wherein the predetermined period includes at least three days.

58. The method according to claim 44, further comprising the step of stopping the upward titration of basal insulin if, after titrating the recommended dose over a predetermined period, the minimum glucose metric does not decrease.

59. The method according to claim 44, further comprising the step of stopping the downward titration of basal insulin if the minimum glucose metric does not increase after titrating the recommended dose over a predetermined period of time.

60. The method according to claim 44, further comprising the step of activating a blind mode so that glucose data is not displayed to the user.

61. If the change in the minimum glucose metric is outside the predetermined metric, the step is to start a counter configured to monitor the number of titration cycles. The method according to claim 44, further comprising the step of recommending the cessation of basal insulin titration if, after a predetermined number of titration cycles in which the dose of basal insulin is increased or decreased, the change in the minimum glucose metric remains outside the predetermined metric.

62. The method according to claim 44, further comprising the step of outputting a prompt to the user instructing the user to seek guidance from a medical professional.

63. The method according to claim 44, further comprising the step of outputting a quiz to the user in order to verify that the user is following the recommended dosage.

64. A method for detecting non-adherence to dose recommendations, wherein the method is: The process involves receiving the user's glucose data from an in vivo glucose monitoring device during the first period, A step of determining a first minimum glucose metric for the first period, A step of providing the user with a dosage recommendation based on the glucose data during the first period, The process of receiving the user's glucose data during a second period following the dose recommendation, A step of determining a second minimum glucose metric for the second period, A step of determining non-adherence to the recommended dosage based on a comparison of the first minimum glucose metric and the second minimum glucose metric, A method comprising the step of outputting an indicator of non-adherence.

65. The method according to claim 64, wherein the first minimum glucose metric and the second minimum glucose metric include the minimum hourly average glucose (DMHAG) of a day.

66. The method according to claim 64, wherein the comparison includes a change between the first minimum glucose metric and the second minimum glucose metric.

67. The method according to claim 66, wherein the step of determining non-adherence includes comparing the change between the first minimum glucose metric and the second minimum glucose metric with a predetermined percentage change threshold.

68. The method according to claim 66, wherein the step of determining non-adherence includes comparing the change between the first minimum glucose metric and the second minimum glucose metric with a predetermined change in blood glucose level.

69. The method according to claim 64, wherein the comparison includes directions for the first minimum glucose metric and the second minimum glucose metric.

70. The method according to claim 69, wherein the direction is inversely proportional to the recommended dose.

Images