Systems and methods for providing therapy management guidance related to diabetes for health improvement

Abstract

Certain aspects of the present disclosure provide a therapy management system configured to monitor at least glucose concentration levels of a patient during a first time period using one or more glucose sensors to obtain glucose data, and determine a diabetic state classification of the patient based on the glucose data. The diabetic state classification indicates a diabetic state of the patient and comprises one of: a healthy classification, a pre-diabetic classification, a mild type 2 diabetes (T2D) classification, a severe T2D classification, or a gestational diabetes classification. The therapy management system is further configured to provide personalized feedback to the patient based on the determined diabetic state classification, wherein the personalized feedback comprises a recommendation for at least one of an exercise session, a dietary change, a lifestyle modification, a medication adjustment, or a continuous glucose monitoring wear schedule.

Description

Attorney Docket No.: 0917-PCT01SYSTEMS AND METHODS FOR PROVIDING THERAPY MANAGEMENT GUIDANCE RELATED TO DIABETES FOR HEALTH IMPROVEMENT CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Patent Application No.63 / 739,467, filed December 27, 2024, which is incorporated by reference herein in its entirety, and is hereby expressly made a part of this specification.BACKGROUND

[0002] Dysregulation of a patient’s metabolism and impaired clearance of glucose can lead to Type 2 Diabetes (T2D), pre-diabetes, and / or gestational diabetes and other diseases. The progression in the severity of diabetes from pre-diabetes to T2D, for example, manifests as worsening glucose control over time and an increase in the body’s insulin resistance. Various classifications of the diabetes disease stage (“diabetic state”) of a patient exist based on the severity of the disease, which can be due to the presence of insulin in the patient’s body and / or the ability of the patient to appropriately utilize insulin to clear glucose from the body. Diabetes can be classified based on stage, such as pre-diabetes, T2D, and gestational diabetes. Each stage can be further classified by root cause and severity. Root cause classification can include insulin resistant T2D, insulin deficient T2D, obesity-related T2D, and age-related T2D. T2D may further be classified as mild or severe across one or more of the above root causes. Further, pre-diabetes may include insulin resistant pre-diabetes and insulin deficient pre-diabetes, and gestational diabetes may include insulin resistant gestational diabetes or insulin deficient gestational diabetes in the second trimester or the third trimester.

[0003] Despite such differences, typically, treatment for glucose dysregulation in the form of T2D, pre-diabetes, or gestational diabetes appear to be general and not customized to the array of different stages, and root causes for dysregulation of the metabolism and the impaired clearance of glucose that contribute to a patient’s diabetic state.BRIEF DESCRIPTION OF THE DRAWINGS

[0004] So that the manner in which the above-recited features of the present disclosure can be understood in detail, a more particular description, briefly summarized above, may be had byAttorney Docket No.: 0917-PCT01reference to aspects, some of which are illustrated in the drawings. It is to be noted, however, that the appended drawings illustrate only certain typical aspects of this disclosure and are therefore not to be considered limiting of its scope, for the description may admit to other equally effective aspects.

[0005] FIG. 1 illustrates aspects of an example therapy management system used in connection with implementing embodiments of the present disclosure.

[0006] FIG.2 is a diagram conceptually illustrating an example continuous analyte monitoring system including example continuous analyte sensor(s) with sensor electronics, according to certain embodiments of the present disclosure.

[0007] FIG. 3 illustrates example inputs and example metrics that are calculated based on the inputs for use by the therapy management system of FIG. 1, according to certain embodiments of the present disclosure.

[0008] FIG.4 is a flow diagram illustrating an example method for classifying a diabetic state of a patient, providing personalized therapy management guidance for managing the patient’s health based on the patient’s diabetic state, and monitoring the progression and / or regression of the patient’s diabetic state over time, according to certain embodiments of the present disclosure.

[0009] FIG. 5 is a flow diagram an example method for providing therapy management guidance to patients with various diabetic state classifications and monitoring the progression and / or regression of the patient’s diabetic state over time, according to certain embodiments of the present disclosure.

[0010] FIG. 6 is a flow diagram depicting a method for training machine learning models to classify a patient within a specific diabetic state classifications, provide therapy management guidance to a patient to manage the patient’s health based on the patient’s classification, and / or monitor the progression and / or regression of the patient’s diabetic state over time according to certain embodiments of the present disclosure.

[0011] FIG. 7 is a block diagram depicting a computing device configured to perform the operations of FIGs. 4 and 5, according to certain embodiments of the present disclosure.

[0012] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures. It is contemplated thatAttorney Docket No.: 0917-PCT01elements disclosed in one aspect may be beneficially utilized on other aspects without specific recitation.DETAILED DESCRIPTION

[0013] Accurate classification of a patient’s diabetes, or diabetic dysglycemia is important to ensure the patient is properly monitored over time and that appropriate treatment is recommended to the patient to manage the patient’s diabetes, or diabetic dysglycemia and prevent further progression of the disease. That is, the stage (e.g., prediabetes, moderate diabetes, severe diabetes), type of diabetes (e.g., gestational diabetes), and underlying cause (e.g., insulin resistance, insulin deficiency, genetic root causes, metabolic dysfunction, hormonal fluctuations, etc.) that leads to the dysglycemic responses of the patient are important to ensure proper treatment. As an example, medications prescribed for raising insulin sensitivity are only helpful for a patient, if the patient’s diabetic state stage and root cause indicate that the patient’s dysglycemic glucose response is due to a specific level of insulin resistance, whereas, for other patients with insulin deficiency or more severe stages where insulin resistance cannot be treated a different medication (e.g., insulin therapy) might be better suited. Similarly, depending on the stage of the diabetic state of the patient as well as the type of underlying root cause (e.g., insulin resistance, insulin deficiency, genetic root causes, metabolic dysfunction, etc.), different variations of meal modifications and exercise may be effective. For example, the time of day at which the patient exercises, and the type of exercise is dependent on these factors. Furthermore, whether diet modifications, or exercise modifications should be used as primary means of treatment are dependent on these factors. Because compliance to treatment is likely to improve with more effective and personalized treatment, classifying the patient’s glycemic profile is important to ensure proper treatment that is personalized according to the type of diabetes, stage of diabetes and the root cause of the dysglycemia can help with ensuring that the patient is provided the right treatment. Similarly, by having personalized detection of these factors, the patient can receive more specific alarms and alerts, instead of alarms and alerts for generic events that may not be as useful to the patient. This provides the added advantage of protecting against alarm nuisance and improving patient safety. Additionally, by classifying the patient and understanding the correct metrics and treatments that are personalized for that user, the system can require less processing time and resources because not all metrics need to be generated, and alarms and treatment notifications can be curated to remove unneeded processing and power consumption.Attorney Docket No.: 0917-PCT01

[0014] However, currently, the most common means for monitoring T2D, pre-diabetes, and gestational diabetes patients usually takes place during a doctor’s visit. In other words, for example, a patient who is at risk of diabetes may visit their physician every few months and perform a single point in time glucose test, based on which the patient may be classified as a patient with pre-diabetes, gestational diabetes, or T2D. However, single point in time measurements performed every few months provide a very limited snapshot of the patient’s glucose control, which does very little when it comes to classifying the patient. Further, because the blood test and / or doctor visits are few and far between, it is not technically possible to monitor the patient’s change in glucose control, as well as insulin production and severity of glucose dysregulation, in a timely manner. Even further, single point in time glucose measurements may be influenced by the patient’s diet and / or an activity performed by the patient prior to the measurement, which may lead to a misdiagnosis of the patient.

[0015] Furthermore, a diabetes diagnosis is typically based on difficult and / or unreliable tests (oral glucose tolerance tests or A1C tests). These tests may lead to late diagnosis or misdiagnosis of diabetes, including under diagnosis and over diagnosis. Even further, when a patient is diagnosed, various treatments are prescribed on a general population level rather than being personalized, while there is no way of monitoring adherence to the treatment until the patient’ s next doctor’ s visit. At the patient’ s next doctor’ s visit, adherence would be determined again based on single point in time tests and there is no way of assessing which treatments were followed and lead to results if any

[0016] As a result of the technical deficiencies described above, individuals can experience worsening glucose control without their knowledge. Worsening glucose control can lead to the development of pre-diabetes, T2D, and / or gestational diabetes, the progression of the patient’s diabetic state (e.g., worsening severity of diabetes), and / or the development of insulin deficiency and / or insulin resistance over time. On the other hand, if a patient is informed of worsening glucose control prior to the development or further progression of the patient’s diabetic state and properly classified within a specific diabetic state classification, the patient may be prescribed lifestyle changes, treatments, and / or medical intervention specific to the patient’s diabetic state to improve the patient’s glucose control and diabetes outcome and the adherence and efficacy of such treatments may be assessed. Consequently, there is a need to provide a continuous, cost-effective, and less intrusive analyte monitoring system to help guide healthcare actions for patients withAttorney Docket No.: 0917-PCT01T2D, gestational diabetes, and / or pre-diabetes in a personalized, real time manner, that leads to patient safety and reduced processing and power consumption.

[0017] Accordingly, certain embodiments described herein provide a technical solution to the technical problem described above by providing a continuous analyte monitoring system, including, for example, a continuous glucose sensor for use in performing one or more of accurately classifying a patient’s specific diabetic state, providing therapy management guidance to manage the patient’s health based on the patient’s diabetic state, monitoring adherence to such therapy management guidance and monitoring the progression and / or regression of the patient’s diabetic state over time.

[0018] As used herein, the term “continuous” analyte monitoring refers to monitoring one or more analytes in a fully continuous, semi-continuous, periodic manner, which results in a data stream of analyte values over time. A data stream of analyte values over time is what allows for meaningful data and insight to be derived using the algorithms described herein for determining a patient’s specific diabetic state, providing therapy management guidance to manage the patient’s health based on the patient’s diabetic state, and / or monitoring the progression and / or regression of the patient’s diabetic state over time. A patient’s diabetic state as used herein, refers to a glycemic profile for the patient generated based on monitoring one or more glucose metrics for the patient. The glucose profile can include a first classification of the type of diabetes for a patient, e.g., type 1, type 2, prediabetes, metabolic dysfunction, etc. The glucose profile can additionally include a stage of disease for the patient, such as moderate or severe dysglycemia. Such staging can for example be based on predetermined thresholds. The thresholds can be generated dynamically and personalized for a patient or a specific population of patients, can be based on population based analysis, and / or can take into account clinical thresholds derived based on standards of care or clinical studies in the field. The thresholds can include for example, determination of a insulin deficiency level, insulin resistance level, metabolic flexibility, or other similar metrics using analyte data, such as glucose data, and other analytes described herein for the patient. The glucose profile can also include one or more underlying causes for the diabetic state of the patient. For example, the glucose profile can include whether the patient is insulin resistant, insulin deficient, has metabolic dysfunction or other comorbid conditions that lead to dysglycemia. The glucose profile may result in one or more scores or metrics that help provide the patient with a clear indication of the patient’s diabetic state, as well as personalized treatment and insights (e.g.,Attorney Docket No.: 0917-PCT01alarms / alerts, or other notifications) to ensure the patient is receiving pertinent feedback regarding their glycemic patterns. The term diabetic state is used herein to mean generally the glycemic state of the patient, and does not exclude patients with other glycemic dysregulation conditions that may not traditionally be classified as diabetes.

[0019] In other words, single point in time measurements collected as a result of a patient visiting their health care professional every few months results in sporadic data points (e.g., that are, at best, months apart in timing) that cannot form the basis of any meaningful data or insight to be derived. As such, without the continuous analyte monitoring system of the embodiments herein, it is simply impossible to continuously monitor a patient’s diabetic state, continuously provide therapy management guidance to manage a patient’s health, and continuously monitor the progression and / or regression of the patient’s diabetic state over time, as described herein.

[0020] Further, the data stream of analyte values collected over time, with the continuous analyte monitoring system presented herein, include real-time analyte values, which allows for deriving meaningful data and insight in real-time using the systems and algorithms described herein. The derived real-time data and insight in turn allows for providing real-time determination of a patient’s specific diabetic state, therapy management guidance to manage the patient’s health based on the patient’s diabetic state, and / or monitoring of the progression and / or regression of the patient’s diabetic state over time. Real time analyte values herein refer to analyte values that become available and actionable within seconds or minutes of being produced as a result of at least one sensor electronics module of the continuous analyte monitoring system (1) converting sensor current(s) (i.e., analog electrical signals) generated by the continuous analyte sensor(s) into sensor count values, (2) calibrating the count values to generate at least glucose concentration values using calibration techniques described herein to account for the sensitivity of the continuous analyte sensor(s), and (3) transmitting measured glucose concentration data, including glucose concentration values, to a display device via wireless connection.

[0021] For example, the at least one sensor electronics module is configured to sample the analog electrical signals at a particular sampling period (or rate), such as every 1 second (1 Hz), 5 seconds, 10 seconds, 30 seconds, 1 minute, 3 minutes, 5 minutes, etc., and to transmit the measured glucose concentration data to a display device at a particular transmission period (or rate), whichAttorney Docket No.: 0917-PCT01may be the same as (or longer than) the sampling period, such as every 1 minute (0.016 Hz), 5 minutes, 10 minutes, etc.

[0022] The real-time analyte data that is continuously generated by the continuous analyte monitoring system described herein, therefore, allows the therapy management system herein to determine a patient’s specific diabetic state, provide therapy management guidance to manage the patient’s health based on the patient’s diabetic state, and / or monitor the progression and / or regression of the patient’s diabetic state over time, in real-time, which is technically impossible to perform using existing or conventional techniques or systems. Further, because of the real-time nature of this data, it is also humanly impossible to continuously manually / mentally process a realtime data stream of analyte values over time to derive meaningful data and insight using the algorithms and systems described herein for determining a patient’s specific diabetic state, providing therapy management guidance to manage the patient’s health based on the patient’s diabetic state, and / or monitoring the progression and / or regression of the patient’s diabetic state over time.

[0023] In other words, deriving meaningful data and insight from a stream of real-time data that is continuously generated, processed, calibrated, and analyzed, using the algorithms and systems described herein, is not a task that can be mentally performed. For example, executing the algorithm described in relation to FIGs. 4 and 5 in real-time and on a continuous basis, which would involve using a stream of real-time data that is continuously generated by a patient’s continuous analyte monitoring system and / or significantly large amount of population data (e.g., hundreds or thousands of data points for each one of thousands or millions of patients in the patient population) is not a task that can be mentally performed, especially in real-time at times.

[0024] Further, certain embodiments herein are directed to a technical solution to a technical problem associated with analyte sensor systems. For example, the therapy management system described herein maximizes an accuracy of a classification of a patient’s specific diabetic state and provides therapy management guidance in view of such classification, where such therapy management guidance includes automatically implementing one or more device settings (e.g., thresholds, diet and exercise schedules, etc.) within the therapy management system. In this way, adjustments to the therapy management system settings by the patient may be minimized, which also minimizes device hardware computation and / or network load requirements associated withAttorney Docket No.: 0917-PCT01those adjustments. When this process is implemented for a large group of patients, automatic classification and therapy management guidance will significantly reduce network and / or computation requirements for the group, thereby improving performance of the one or more hardware computing systems implementing such therapy management systems.

[0025] Further, by accurately classifying a patient’ s diabetic state using the analyte monitoring system and providing therapy management guidance, such as treatment recommendations, based on this classification, an accuracy of such treatment recommendations may be improved. This improved accuracy may in turn improve medication dosing instructions (e.g., dosing instructions sent to a hardware insulin pump) as well as treatment recommendations sent to the patient by the therapy management system. Improved recommendations (such as diet, exercise, and medication recommendations) provided by the therapy management system may be followed by the patient, resulting in a favorable improvement of the patient’s analyte data and overall health.

[0026] Additionally, as analyte data of the patient is continuously received over time, the therapy management system may identify the results of earlier therapy management guidance (both for a current patient as well as other patients sharing one or more characteristics with the current patient) and may continually refine future therapy management guidance for the current patient and other related patients based at least in part on these results. The continuous refinement of future therapy management guidance may improve the accuracy of guidance generated by the therapy management system for all patients.

[0027] Additionally, each analyte sensor system that is manufactured by a sensor manufacturer might perform slightly different. As such, there might be inconsistencies between sensors and the measurements they generate once in use. Accordingly, certain embodiments herein are directed to determining the performance of an analyte sensor system during a manufacturing calibration process (in vitro), which includes quantifying certain sensor operating parameters, such as a calibration slope (also known as calibration sensitivity), a calibration baseline, etc.

[0028] Generally, calibration sensitivity refers to the amount of electrical current produced by an analyte sensor of an analyte sensor system when immersed in a predetermined amount of a measured analyte. The amount of electrical current may be expressed in units of picoAmps (pA) or counts. The amount of measured analyte may be expressed as a concentration level in units of milligrams per deciliter (mg / dL), and the calibration sensitivity may be expressed in units ofAttorney Docket No.: 0917-PCT01pA / (mg / dL) or counts / (mg / dL). The calibration baseline refers to the amount of electrical current produced by the analyte sensor when no analyte is detected, and may be expressed in units of pA or counts.

[0029] The calibration sensitivity, calibration baseline, and other information related to the sensitivity profile for the analyte sensor system may be programmed into the sensor electronics module of the analyte sensor system during the manufacturing process, and then used to convert the analyte sensor electrical signals into measured analyte concentration levels. For example, the calibration slope (calibration sensitivity) may be used to predict an initial in vivo sensitivity (Mo) and a final in vivo sensitivity (Mf), which are programmed into the sensor electronics module and used to convert the analyte sensor electrical signals into measured analyte concentration levels.

[0030] In certain embodiments, during in vivo use, the sensor electronics module of an analyte sensor system samples the analog electrical signals produced by the analyte sensor to generate analyte sensor count values, and then determines the measured analyte concentration levels based on the analyte sensor count values, the initial in vivo sensitivity (Mo), and the final in vivo sensitivity (Mf). For example, measured analyte concentration levels may be determined using a sensitivity function M(t) that is based on the initial in vivo sensitivity (Mo) and the final in vivo sensitivity (Mr). The sensitivity function M(t) may expressed in several different ways, such as a simple correction factor that is not dependent on elapsed time (ti) of in vivo use, a linear relationship between sensitivity and time (ti), an exponential relationship between sensitivity and time (ti), etc. Equation 1 presents one technique for determining a measured analyte concentration level (ACL) from an analyte sensor count value (count) at a time h:ACL = count / M(ti) Eq. 1 A calibration baseline (baseline) may also be used to determine a measured analyte concentration level (ACL) from an analyte sensor count value (count) at a time ti, and Equation 2 presents one technique:ACL = (count - baseline) / M(ti) Eq. 2Example Therapy Management System Including an Example Analyte Sensor for Providing Therapy Management Support to Patients having Various Classifications of Diabetic StateAttorney Docket No.: 0917-PCT01

[0031] FIG.l illustrates an example therapy management system 100 for classifying a diabetic state of a patient 102 and / or providing therapy management guidance to optimize the patient’s health based on the diabetic state of the patient, using a continuous analyte monitoring system 104 configured to continuously measure the patient’s glucose levels. A patient, in certain embodiments, is a user for which a continuous analyte sensor provides measurements of the patient’s analyte (i.e., a patient wearing the continuous analyte sensor). The patient can include a healthy patient, or a patient with any diabetes or a comorbid condition, or dysglycemia including a patient with T2D, a patient with pre-diabetes, a patient with gestational diabetes, and / or a patient at risk of developing T2D, for example.

[0032] In certain embodiments, therapy management system 100 includes continuous analyte monitoring system 104, a display device 107 that executes application 106, a therapy management engine 114, a patient database 110, a historical records database 112, a training server system 140, and a therapy management engine 114, each of which is described in more detail below.

[0033] The term “analyte” as used herein is a broad term used in its ordinary sense, including, without limitation, to refer to a substance or chemical constituent in a biological fluid (for example, blood, interstitial fluid, cerebral spinal fluid, lymph fluid or urine) that can be analyzed. Analytes can include naturally occurring substances, artificial substances, metabolites, and / or reaction products. Analytes for measurement by the devices and methods include, but are not limited to, potassium, glucose, endogenous insulin, acarboxyprothrombin; acylcarnitine; exogenous insulin; adenine phosphoribosyl transferase; adenosine deaminase; albumin; albumin-creatinine ratio; alpha-fetoprotein; amino acid profiles (arginine (Krebs cycle), histidine / urocanic acid, homocysteine, phenylalanine / tyrosine, tryptophan); androstenedione; antipyrine; arabinitol enantiomers; arginase; benzoylecgonine (cocaine); biotinidase; biopterin; c-peptide; c-reactive protein; carnitine: camosinase; CD4; ceruloplasmin; chenodeoxycholic acid; chloroquine; cholesterol; cholinesterase; conjugated 1-P hydroxy-cholic acid; cortisol; creatine kinase; creatine kinase MM isoenzyme; creatinine; cyclosporin A; cystatin C; d-penicillamine; deethylchloroquine; dehydroepiandrosterone sulfate; DNA (acetylator polymorphism, alcohol dehydrogenase, alpha 1 -antitrypsin, glucose-6-phosphate dehydrogenase, hemoglobin A, hemoglobin S, hemoglobin C, hemoglobin D, hemoglobin E, hemoglobin F, D-Punjab, hepatitis B virus, HCMV, HIV-1, HTLV-1, MCAD, RNA, PKU, Plasmodium vivax, 21-deoxycortisol); desbutylhalofantrine; dihydropteridine reductase; diptheria / tetanus antitoxin; erythrocyteAttorney Docket No.: 0917-PCT01arginase; erythrocyte protoporphyrin; esterase D; fatty acids / acylglycines; free P-human chorionic gonadotropin; free erythrocyte porphyrin; free thyroxine (FT4); free tri-iodothyronine (FT3); fumarylacetoacetase; galactose / gal- 1 -phosphate; galactose- 1 -phosphate uridyltransferase; gentamicin; glucose-6-phosphate dehydrogenase; glutathione; glutathione perioxidase; glycocholic acid; glycosylated hemoglobin; halofantrine; hemoglobin variants; hexosaminidase A; human erythrocyte carbonic anhydrase I; 17-alpha-hydroxyprogesterone; hypoxanthine phosphoribosyl transferase; immunoreactive trypsin; lactate; pyruvate; lead; lipoproteins ((a), B / A-l, ); lysozyme; mefloquine; netilmicin; phenobarbitone; phenytoin; phytanic / pristanic acid; progesterone; prolactin; prolidase; purine nucleoside phosphorylase; proteinuria; quinine; reverse tri-iodothyronine (rT3); selenium; serum pancreatic lipase; sisomicin; somatomedin C; specific antibodies recognizing any one or more of the following that may include (adenovirus, anti-nuclear antibody, anti-zeta antibody, arbovirus, Aujeszky's disease virus, dengue virus, Dracunculus medinensis, Echinococcus granulosus. Entamoeba histolytica, enterovirus, Giardia duodenalisa, Helicobacter pylori, hepatitis B virus, herpes virus, HIV-1, IgE (atopic disease), influenza virus, Leishmania donovani, leptospira, measles / mumps / rubella, melatonin, Mycobacterium leprae, Mycoplasma pneumoniae, Myoglobin, Onchocerca volvulus, parainfluenza virus, Plasmodium falciparum, poliovirus, Pseudomonas aeruginosa, pro-C3, respiratory syncytial virus, rickettsia (scrub typhus), Schistosoma mansoni, Toxoplasma gondii, Trepenoma pallidium, Trypanosoma cruzi / rangeli. vesicular stomatis virus, Wuchereria bancrofti, yellow fever virus); specific antigens (hepatitis B virus, HIV-1); succinylacetone; sulfadoxine; theophylline; thyrotropin (TSH); thyroxine (T4); thyroxine-binding globulin; trace elements; transferrin; UDP-galactose-4-epimerase; urea; uroporphyrinogen I synthase; vitamin A; white blood cells; and zinc protoporphyrin.

[0034] Salts, sugar, protein, fat, vitamins, and hormones (e.g., insulin) naturally occurring in blood or interstitial fluids can also constitute analytes in certain implementations. The analyte can be naturally present in the biological fluid, for example, a metabolic product, a hormone, an antigen, an antibody, and the like. Alternatively, the analyte can be introduced into the body or exogenous, for example, a contrast agent for imaging, a radioisotope, a chemical agent, a fluorocarbon -based synthetic blood, or a drug or pharmaceutical composition, including but not limited to insulin; glucagon, ethanol; cannabis (marijuana, tetrahydrocannabinol, hashish); inhalants (nitrous oxide, amyl nitrite, butyl nitrite, chlorohydrocarbons, hydrocarbons); cocaineAttorney Docket No.: 0917-PCT01(crack cocaine); stimulants (amphetamines, methamphetamines, Ritalin, Cylert, Preludin, Didrex, PreState, Voranil, Sandrex, Plegine); depressants (barbiturates, methaqualone, tranquilizers such as Valium, Librium, Miltown, Serax, Equanil, Tranxene); hallucinogens (phencyclidine, lysergic acid, mescaline, peyote, psilocybin); narcotics (heroin, codeine, morphine, opium, meperidine, Percocet, Percodan, Tussionex, Fentanyl, Darvon, Talwin, Lomotil); designer drugs (analogs of fentanyl, meperidine, amphetamines, methamphetamines, and phencyclidine, for example, Ecstasy); anabolic steroids; and nicotine. The metabolic products of drugs and pharmaceutical compositions are also contemplated analytes. Analytes such as neurochemicals and other chemicals generated within the body can also be analyzed, such as, for example, ascorbic acid, uric acid, dopamine, noradrenaline, 3-methoxytyramine (3MT), 3,4-Dihydroxyphenylacetic acid (DOPAC), Homovanillic acid (HVA), 5-Hydroxytryptamine (5HT), and 5-Hydroxyindoleacetic acid (FHIAA), and intermediaries in the Citric Acid Cycle.

[0035] While certain embodiments herein involve measuring and analyzing glucose using the systems and methods described herein, in some other embodiments, other analytes, such as lactate, ketones, potassium, creatinine, amino acids, and FFAs, or other analytes listed above are additionally or alternatively used in accurately classifying a patient’s diabetic state, providing therapy management guidance to manage the patient’s health based on the patient’s diabetic state, and / or monitoring the progression and / or regression of the patient’s diabetic state over time.

[0036] In certain embodiments, continuous analyte monitoring system 104 is configured to continuously measure one or more analytes. The analyte measurements (either in raw for, or a processed form that represent the measurements) can be transmitted via the analyte monitoring system 104 or the display device 107 to an electronic medical records (EMR) system (not shown in FIG. 1). An EMR system is a software platform which allows for the electronic entry, storage, and maintenance of digital medical data. An EMR system is generally used throughout hospitals and / or other caregiver facilities to document clinical information on patients over long periods. EMR systems organize and present data in ways that assist clinicians with, for example, interpreting health conditions and providing ongoing care, scheduling, billing, and follow up. Data contained in an EMR system may also be used to create reports for clinical care and / or disease management for a patient. In certain embodiments, the EMR may be in communication with therapy management engine 114 (e.g., via a network) for performing the techniques described herein. In particular, as described herein, therapy management engine 114 may obtain dataAttorney Docket No.: 0917-PCT01associated with a patient, use the obtained data as input into one or more trained model(s), and output a prediction. In some cases, the EMR provides the data to therapy management engine 114 to be used as input into the one or more models. Further, in some cases, therapy management engine 114, after making a prediction, provides the output prediction to the EMR.

[0037] In certain embodiments, continuous analyte monitoring system 104 is configured to continuously measure one or more analytes and transmit the analyte measurements to display device 107 for use by application 106. In some embodiments, continuous analyte monitoring system 104 transmits the analyte measurements to display device 107 through a wireless connection (e.g., Bluetooth connection, WiFi connection, local area network connection, etc.). In certain embodiments, display device 107 is a smart phone. However, in certain other embodiments, display device 107 is any other type of computing device such as a laptop computer, a smart watch, a tablet, a standalone receiver, or any other computing device capable of executing application 106. In some embodiments, continuous analyte monitoring system 104 and / or analyte sensor application 106 transmit the analyte measurements to one or more other individuals having an interest in the health of the patient (e.g., a family member or physician for real-time treatment and care of the patient). Continuous analyte monitoring system 104 is described in more detail with respect to FIG.2.

[0038] Application 106 is a mobile health application that is configured to receive and analyze analyte measurements from continuous analyte monitoring system 104. In particular, application 106 stores information about a patient, including the patient’s analyte measurements, in a patient profile 118 associated with the patient for processing and analysis, as well as for use by therapy management engine 114 to provide therapy management guidance to the patient.

[0039] Therapy management engine 114 refers to a set of software instructions with one or more software modules, including data analysis module (DAM) 116. In certain embodiments, therapy management engine 114 executes entirely on one or more computing devices in a private or a public cloud. In such embodiments, application 106 communicates with therapy management engine 114 over a network (e.g., Internet). In some other embodiments, therapy management engine 114 executes partially on one or more local devices, such as display device 107 and / or continuous analyte monitoring system 104, and partially on one or more computing devices in a private or a public cloud. In some other embodiments, therapy management engine 114 executesAttorney Docket No.: 0917-PCT01entirely on one or more local devices, such as display device 107 and / or continuous analyte monitoring system 104. As discussed in more detail herein, therapy management engine 114 can provide therapy management guidance to the patient via application 106 for optimal monitoring system wear recommendations, diet recommendations, mealtimes, exercise regimen, medication recommendations, medication dose recommendations, etc. to manage a patient’s health based on the patient’s diabetic state and / or information included in patient profile 118.

[0040] Patient profile 118 may include information collected about the patient from application 106. For example, application 106 provides a set of inputs 130, including the analyte measurements received from continuous analyte monitoring system 104, that are stored in patient profile 118. In certain embodiments, inputs 130 provided by application 106 include other data in addition to analyte measurements received from continuous analyte monitoring system 104. For example, application 106 can obtain additional inputs 130 through manual patient input, one or more other non-analyte sensors or devices, other applications executing on display device 107, etc. Non-analyte sensors and devices include one or more of, but are not limited to, an insulin pump sensor, stretch sensor, body sound sensor, acoustic gastography sensor, an electrocardiogram (ECG) sensor, a heart rate monitor, a blood pressure sensor, a respiratory sensor, a thermometer, a digital weight scale, sensors or devices provided by display device 107 (e.g., accelerometer, camera, global positioning system (GPS), heart rate monitor, etc.), or other patient accessories (e.g., a smart watch or fitness tracker), or any other sensors or devices that provide relevant information about the patient. Inputs 130 of patient profile 118 provided by application 106 are described in further detail below with respect to FIG. 3.

[0041] DAM 116 of therapy management engine 114 is configured to process the set of inputs 130 to determine one or more metrics 132. Metrics 132, discussed in more detail below with respect to FIG. 3. may, at least in some cases, be generally indicative of the disease state of a patient, such as one or more of the patient’s general analyte trends, trends associated with the health of the patient, etc. In certain embodiments, metrics 132 can then be used by therapy management engine 114 as input for classifying a patient’s diabetic state, providing therapy management guidance to manage the patient’s health based on the patient’s classification, and / or monitoring the progression and / or regression of the patient’s diabetic state over time. As shown, metrics 132 are also stored in patient profile 118.Attorney Docket No.: 0917-PCT01

[0042] Patient profile 118 also includes demographic info 120, physiological info 122, disease info 124, and / or medication info 126. In certain embodiments, such information can be provided through patient input, obtained from one or more analyte or non-analyte sensors, or obtained from certain data stores (e.g., electronic medical records (EMRs), etc.). In certain embodiments, demographic info 120 includes one or more of the patient’s age, ethnicity, gender, etc. In certain embodiments, physiological info 122 includes one or more of the patient’s height, weight, and / or body mass index (BMI). In certain embodiments, disease info 124 includes information about a disease of a patient, such diagnoses of type 1 or type 2 diabetes, and / or other conditions related to diabetes (e.g., liver disease and / or liver dysfunction, kidney disease and / or chronic kidney disease, heart disease, obesity), etc. In certain embodiments, information about a patient’s disease information also includes the length of time since diagnosis, the level of disease control, level of compliance with disease management therapy, other types of diagnosis (e.g., heart disease, hypertension), or measures of health (e.g., heart rate, exercise, sleep, etc.), and / or the like. In certain embodiments, patient profile 118 includes information about the patient’s drug and / or alcohol consumption.

[0043] In certain embodiments, medication info 126 includes information about the amount, frequency, and type of a medication taken by a patient. In certain embodiments, the amount, frequency, and type of a medication taken by a patient is time-stamped and correlated with the patient’s analyte levels, thereby, indicating the impact the amount, frequency, and type of the medication had on the patient’s analyte levels.

[0044] In certain embodiments, medication information includes information about consumption of one or more drugs known to treat T2D, one or more drugs known to improve liver function, and / or drugs that may alter the patient’s analyte levels. Examples of drugs known to treat T2D include Metformin, insulin, sulfonylureas, SGLT-2 inhibitors, glucagon-like peptide (GLP-1) receptor agonists, gastric inhibitory polypeptide (GIP) receptor agonists, and the like. For example, exogenous insulin or insulin sensitizers can be utilized to treat T2D by lowering blood glucose levels. Examples of drugs known to improve liver function include nonsteroidal antiinflammatory drugs, antibiotics, birth control pills, antidepressants, opioids, proton-pump inhibitors, and the like. Medication information may further include information on the consumption of steroids, as steroids can undesirably alter a patient’s glucose levels and induce hypoglycemia.Attorney Docket No.: 0917-PCT01

[0045] In certain embodiments, patient profile 118 is dynamic because at least part of the information that is stored in patient profile 118 may be revised over time and / or new information may be added to patient profile 118 by therapy management engine 114 and / or application 106. Accordingly, information in patient profile 118 stored in patient database 110 provides an up-to-date repository of information related to a patient.

[0046] Patient database 110, in some embodiments, refers to a storage server that operates in a public or private cloud. Patient database 110 may be implemented as any type of datastore, such as relational databases, non-relational databases, key-value datastores, file systems including hierarchical file systems, and the like. In some exemplary implementations, patient database 110 is distributed. For example, patient database 110 can comprise a plurality of persistent storage devices, which are distributed. Furthermore, patient database 110 can be replicated so that the storage devices are geographically dispersed.

[0047] Patient database 110 includes patient profiles 118 associated with a plurality of patients who similarly interact with application 106 executing on the display devices 107 of the other patients. Patient profiles stored in patient database 110 are accessible to not only application 106, but therapy management engine 114, as well. Patient profiles in patient database 110 may be accessible to application 106 and therapy management engine 114 over one or more networks (not shown). As described above, therapy management engine 114, and more specifically DAM 116 of therapy management engine 114, can fetch inputs 130 from patient database 110 and compute a plurality of metrics 132 which can then be stored as application data 128 in patient profile 118.

[0048] In certain embodiments, patient profiles 118 stored in patient database 110 can also be stored in historical records database 112. Patient profiles 118 stored in historical records database 112 can provide a repository of up-to-date information and historical information for each patient of application 106. Thus, historical records database 112 essentially provides all data related to each patient of application 106, where data is stored according to an associated timestamp. The timestamp associated with information stored in historical records database 112 can identify, for example, when information related to a patient has been obtained and / or updated.

[0049] Further, historical records database 112 may maintain time series data collected for patients over a period of time, including for patients who use continuous analyte monitoring system 104 and application 106. For example, analyte data for a patient who has used continuous analyteAttorney Docket No.: 0917-PCT01monitoring system 104 and application 106 for a period of time to determine a specific diabetic state, provide therapy management guidance to manage the patient’s health based on the patient’s diabetic state, and / or monitor the progression and / or regression of the patient’s diabetic state over time may have time series analyte data associated with the patient maintained over the period of time. In certain embodiments, the period of time can be 3 days, or 1 week, or one month, or one year, or five years, for example.

[0050] Further, in certain embodiments, historical records database 112 includes data for one or more patients who are not users of continuous analyte monitoring system 104 and / or application 106. For example, historical records database 112 includes information (e.g., patient profile(s)) related to one or more patients with various diabetic state classifications, as well as information (e.g., patient profile(s)) related to one or more patients who have been managing their health based on therapy management guidance (e.g., monitoring system wear recommendations, diet recommendations, and / or exercise recommendations). Data stored in historical records database 112 may be referred to herein as population data, which could include hundreds or thousands of data points for each one of thousands or millions of patients in the patient population. In other words, data stored in historical records database 112 and used in certain embodiments described herein could include gigabytes, terabytes, petabytes, exabytes, etc. of data.

[0051] Data related to each patient stored in historical records database 112 may provide time series data collected over the disease lifetime of the patient. For example, the data includes information about the patient prior to a diabetes diagnosis, including information related to the patient’s historical glucose control, as well as information related to other diseases, such as liver disease and / or liver dysfunction, cardiovascular disease and / or peripheral vascular diseases, kidney disease and / or chronic kidney disease, etc. Such information may indicate symptoms of the patient, physiological states of the patient, analyte levels of the patient, states / conditions of one or more organs of the patient, habits of the patient (e.g., activity levels, food consumption, etc.), medication prescribed, etc. over a period of time.

[0052] Although depicted as separate databases for conceptual clarity, in some embodiments, patient database 110 and historical records database 112 can operate as a single database. That is, historical and current data related to patients of continuous analyte monitoring system 104 and application 106, as well as historical data related to patients that were not previously patients ofAttorney Docket No.: 0917-PCT01continuous analyte monitoring system 104 and application 106, can be stored in a single database. The single database can be a storage server that operates in a public or private cloud.

[0053] As mentioned previously, therapy management system 100 is configured to determine a patient’s specific diabetic state, provide therapy management guidance to manage the patient’s health based on the patient’s diabetic state, and / or monitor the progression and / or regression of the patient’s diabetic state over time using continuous analyte monitoring system 104. In certain embodiments, therapy management engine 114 is configured to provide real-time and or non-real-time therapy management guidance to the patient and / or others, including but not limited, to healthcare providers, family members of the patient, caregivers of the patient, researchers, artificial intelligence (Al) engines, and / or other individuals, systems, and / or groups supporting care or learning from the data.

[0054] For example, therapy management engine 114 can be used to collect information associated with a patient in patient profile 118 stored in patient database 110. to perform analytics thereon for classifying a diabetic state of a patient, providing therapy management guidance to the patient based on the patient’s diabetic state, and / or monitoring the progression and / or regression of the patient’s diabetic state over time. Patient profile 118 may be accessible to therapy management engine 114 over one or more networks (not shown) for performing such analytics.

[0055] In certain embodiments, therapy management engine 114 utilizes one or more trained machine learning models capable of providing a diabetic state classification and corresponding therapy management guidance based on information that therapy management engine 114 has collected and / or received from patient profile 118. In the illustrated embodiment of FIG. 1, therapy management engine 114 utilizes trained machine learning model(s) provided by a training server system 140. Although depicted as a separate server for conceptual clarity, in certain embodiments, training server system 140 and therapy management engine 114 operate as a single server or system. That is, the model is trained and used by a single server and / or system, or is trained by one or more servers and / or systems and deployed for use on one or more other servers and / or systems. In certain embodiments, the model is trained on one or many virtual machines (VMs) running, at least partially, on one or many physical services in relational and or nonrelational database formats.Attorney Docket No.: 0917-PCT01

[0056] Training server system 140 is configured to train the machine learning model(s) using training data, which includes data (e.g., from patient profiles) associated one or more patients (e.g., users or non-users of continuous analyte monitoring system 104 and / or application 106) having various diabetic state classifications and / or managing their diabetes based on the therapy management guidance. The training data may be stored in historical records database 112 and may be accessible to training server system 140 over one or more networks (not shown) for training the machine learning model(s).

[0057] The training data refers to a dataset that has been featurized and labeled. For example, the dataset includes a plurality of data records, each including information corresponding to a different patient profile stored in patient database 110, where each data record is featurized and labeled. In machine learning and pattern recognition, a feature is an individual measurable property or characteristic. Generally, the features that best characterize the patterns in the data are selected to create predictive machine learning models. Data labeling is the process of adding one or more meaningful and informative labels to provide context to the data for learning by the machine learning model.

[0058] As an illustrative example, each relevant characteristic of a patient, which is reflected in a corresponding data record, is a feature used in training the machine learning model. Such features include demographic information (e.g., age, gender, ethnicity, etc.), analyte information (e.g., glucose metrics), non-analyte sensor information (e.g., body sound sensor data, fitness trackers, etc.), medical history and / or disease information (e.g., diabetes, liver disease and / or liver dysfunction, kidney disease and / or chronic kidney disease, heart disease, obesity, etc.), medication information, and / or any other information relevant to classifying a patient’s diabetic state, providing various treatment recommendations to manage the patient’s health based on the patient’s diabetic state, and / or monitoring the progression and / or regression of the patient’s diabetic state over time.

[0059] In addition, the data record is labeled with information the corresponding model is being trained to predict. In one example, if a model is being trained to classify the patient’s diabetic state, then the data records in the training dataset are labeled with different diabetic states. In another example, if a model is being trained to output a therapy management guidance, then the data records in the training dataset are labeled with different therapy management guidance. ForAttorney Docket No.: 0917-PCT01example, if a model is being trained to output therapy management guidance related to likelihood of a specific monitoring system wear period being sufficient to manage a patient’s diabetes, then the data records in the training dataset are labeled with one or more of various monitoring system wear frequencies and corresponding diabetic states. Note that, in one example, such a model is a multi-input single-output (MISO) model, configured to predict only the patient’s diabetic state (e.g., whether the patient has insulin deficient T2D), in which case additional MISO models are trained to each predict the likelihood of a specific monitoring system wear period being sufficient to manage a patient’s diabetes based on their classification, whether the patient would benefit from a specific medication based on their classification, or the like. In another example, such a model is a multi-input multi-output (MIMO) model, configured to provide multiple predictions (e.g., the patient’s diabetic state, the likelihood of a specific monitoring system wear period would be sufficient to manage the patient’s diabetic state based on their classification, whether the patient would benefit from a specific medication based on their classification, etc.).

[0060] The model(s) are then trained by training server and / or system 140 using the featurized and labeled training data. In particular, the features of each data record may be used as input into the machine learning model(s), and the generated output may be compared to label(s) associated with the corresponding data record. The model(s) may compute a loss based on the difference between the generated output and the provided label(s). This loss is then used to modify the internal parameters or weights of the model. By iteratively processing each data record corresponding to each historical patient, the model(s) may be iteratively refined to generate accurate predictions of a classification of a patient’s diabetic state, an optimal recommendation to manage the patient’s health based on the patient’s diabetic state, etc.

[0061] As illustrated in FIG. 1, training server system 140 deploys these trained model(s) to therapy management engine 114 for use during runtime. For example, therapy management engine 114 may obtain patient profile 118 associated with a patient and stored in patient database 110, use information in patient profile 118 as input into the trained model(s), and output a prediction indicative of the patient’s diabetic state, therapy management guidance, including monitoring system wear frequency, for managing the patient’s health based on the patient’s diabetic state, and / or a prediction indicative of the progression and / or regression of the patient’s diabetic state over time (e.g., shown as output 144 in FIG. 1). Output 144 generated by therapy management engine 114 may indicate improvement or deterioration in the patient’s diabetic state and / orAttorney Docket No.: 0917-PCT01effectiveness of different therapy management guidance in managing the patient’s health over time. Output 144 may be provided to the patient (e.g., through application 106), to a caretaker of the patient (e.g., a parent, a relative, a guardian, a teacher, a physical therapist, a fitness trainer, a nurse, etc.), to a physician or healthcare provider of the patient, or any other individual that has an interest in the wellbeing of the patient for purposes of improving the health of the patient, such as, in some cases by effectuating recommended treatment and / or seeking medical intervention. Output 144 generated by therapy management engine 114 is stored in patient database 110 and is utilized to train or re-train the trained model(s) and / or a model-based system.

[0062] In certain embodiments, output 144 generated by therapy management engine 114 may be stored in patient profile 118. Output 144 may be indicative of a patient’s current or future diabetes disease classification, and therapy management guidance for managing the patient’s health based on the patient’s diabetic state, etc. Output 144 stored in patient profile 118 may be continuously updated by therapy management engine 114. Accordingly, for example, classifications of a diabetic state and therapy management guidance, originally stored as outputs 144 in patient profile 118 in patient database 110 and then passed to historical records database 112, may provide an indication of the progression or improvement of the diabetic state classification of a patient over time, as well as provide an indication as to the effectiveness of different therapy management guidance provided to the patient to manage or improve the patient’s health.

[0063] In certain embodiments, a patient’s own historical data is used by training server system 140 to train a personalized model for the patient that provides therapy management guidance and insight around the patient’s medical history / current disease state, average analyte levels, etc. For example, in certain embodiments, a model trained based on population data can be used to provide disease progression feedback to the patient. However, after collecting personalized information (e.g., analyte sensor information, non-analyte sensor information, diabetic state classification, etc.) associated with the patient, the personalized information can be used for further personalizing the model. For example, information obtained over time from the patient can be used to more accurately determine classification of diabetic state, provide personalized therapy management guidance for managing the patient’s health based on the patient’s diabetic state, and monitor progression and / or regression of the patient’s diabetic state over time.Attorney Docket No.: 0917-PCT01

[0064] Further, a patient’s historical data can be used to generate a baseline to indicate progression or regression in the patient’s diabetic state based, for example, on the patient’s analyte metrics (e.g.. baseline, rate of change, minimum and / or maximum levels, post-prandial levels, variability), etc. As an illustrative example, a patient’s data over a time period, such as one week, two weeks ago can be used to generate baseline that can be compared with the patient’ s current data to identify whether the patient’s diabetic state has improved. In certain embodiments, the model can further be able to predict or project out the patient’s diabetic state or their future improvement / deterioration (e.g., worsening or development of diabetes) based on the patient’s recent pattern of data (e.g., analyte data, non-analyte data, medication trends, meal trends, exercise trends, etc.).

[0065] In certain embodiments, historical patient population data based on patients with various diabetic state classifications can be used to generate a baseline to indicate progression or regression in the patient’s diabetic state. In certain other embodiments, known clinical evidence and / or observable data through clinical investigations of procedures can be used to generate a baseline to indicate progression or regression in the patient’s diabetic state.

[0066] In certain embodiments, an AI / ML model can be trained to provide a recommendation for monitoring system wear frequency, medications, diet, exercise, and other types of therapy management recommendations to help the patient manage or improve their diabetic state classification based on the patient’s historical data, including how different types of medication, food and / or activities impacted the patient’s diabetes and / or glucose control. In certain embodiments, an AI / ML model can be trained to make predictions about the underlying cause of certain improvements or deteriorations in the patient’s diabetic state. For example, application 106 can display a user interface with a graph that shows the patient’s analyte levels with trend lines and indicate, e.g., retrospectively, how the body’s analyte levels affected the classification of the patient’s diabetic state and / or glucose control at certain points in time.

[0067] In certain other embodiments, rules-based models can be used alternatively or additionally. For example, a rules-based model can be used to map a patient’s analyte data, non-analyte data, inputs, and / or historical data to certain diabetic state classifications, therapy management guidance for managing the patient’s health based on their diabetic state classification, and / or certain progression and / or regression of diabetic state classification, etc., using, forAttorney Docket No.: 0917-PCT01example, a rules library. In certain embodiments, a rules-based model can map certain inputs to diabetic state classifications, progression and / or regression of diabetic state classification, and / or therapy management guidance for patients with similar inputs in the past.

[0068] FIG. 2 is a diagram 200 conceptually illustrating an example continuous analyte monitoring system 104 including example continuous analyte sensor(s) with sensor electronics, in accordance with certain aspects of the present disclosure. For example, continuous analyte monitoring system 104 can be configured to continuously monitor one or more analytes of a patient, in accordance with certain aspects of the present disclosure.

[0069] Continuous analyte monitoring system 104 in the illustrated embodiment includes sensor electronics module 204 and one or more continuous analyte sensor(s) 202 (individually referred to herein as continuous analyte sensor 202 and collectively referred to herein as continuous analyte sensors 202) associated with sensor electronics module 204. Sensor electronics module 204 can be in wireless communication (e.g., directly or indirectly) with one or more of display devices 210, 220, 230, and 240. In certain embodiments, sensor electronics module 204 can also be in wireless communication (e.g., directly or indirectly) with one or more medical devices, such as medical devices 208 (individually referred to herein as medical device 208 and collectively referred to herein as medical devices 208), and / or one or more other non-analyte sensors 206 (individually referred to herein as non-analyte sensor 206 and collectively referred to herein as non-analyte sensor 206).

[0070] In certain embodiments, a continuous analyte sensor 202 can comprise one or more sensors for detecting and / or measuring analyte(s). The continuous analyte sensor 202 can be a multi-analyte sensor configured to continuously measure two or more analytes or a single analyte sensor configured to continuously measure a single analyte as a non-invasive device, a subcutaneous device, a transcutaneous device, a transdermal device, and / or an intravascular device. In certain embodiments, the continuous analyte sensor 202 can be configured to continuously measure analyte levels of a patient using one or more techniques, such as potentiometric techniques, enzymatic techniques, chemical techniques, physical techniques, electrochemical techniques, spectrophotometric techniques, polarimetric techniques, calorimetric techniques, iontophoretic techniques, radiometric techniques, immunochemical techniques, and the like. The term “continuous.” as used herein, can mean fully continuous, semi-continuous,Attorney Docket No.: 0917-PCT01periodic, etc. In certain aspects, the continuous analyte sensor 202 provides a data stream indicative of the concentration of one or more analytes of the patient. The data stream can include raw data signals, which are then converted into a calibrated and / or filtered data stream used to provide estimated analyte value(s) to the patient.

[0071] In certain embodiments, the continuous analyte sensor 202 can be a multi-analyte sensor, configured to continuously measure multiple analytes in a patient’s body. For example, in certain embodiments, the continuous multi-analyte sensor 202 can be a single sensor configured to measure glucose in the patient’s body.

[0072] In certain embodiments, one or more single-analyte and / or multi-analyte sensors can be used in combination. Information from each of the multi-analyte sensor(s) and / or single analyte sensor(s) can be combined to provide therapy management guidance using methods described herein. In further embodiments, other non-contact and or periodic or semi-continuous, but temporally limited, measurements for physiological information can be integrated into the system such as by including weight scale information or non-contact heart rate monitoring from a sensor pad under the patient while in a chair or bed, through an infra-red camera detecting temperature and / or blood flow patterns of the patient, and / or through a visual camera with machine vision for height, weight, or other parameter estimation without physical contact.

[0073] In certain embodiments, the continuous analyte sensor(s) 202 comprises a percutaneous wire that has a proximal portion coupled to the sensor electronics module 204 and a distal portion with several electrodes, such as a measurement electrode and a reference electrode. The measurement (or working) electrode may be coated, covered, treated, embedded, etc., with one or more chemical molecules that react with a particular analyte, and the reference electrode may provide a reference electrical voltage. The measurement electrode may generate the analog electrical signal, which is conveyed along a conductor that extends from the measurement electrode to the proximal portion of the continuous analyte sensor(s) 202 that is coupled to the sensor electronics module 204. After the continuous analyte monitoring system 104 has been applied to epidermis of the patient, continuous analyte sensor(s) 202 penetrates the epidermis, and the distal portion extends into the dermis and / or subcutaneous tissue under epidermis.

[0074] In certain embodiments, the continuous analyte sensor(s) 202 comprises a planar substrate that has a proximal portion coupled to the sensor electronics module 204 and a distalAttorney Docket No.: 0917-PCT01portion with several electrodes such as a measurement electrode and a reference electrode. The measurement (or working) electrode may be coated, covered, treated, embedded, etc., with one or more chemical molecules that react with a particular analyte, and the reference electrode may provide a reference electrical voltage. The measurement electrode may generate the analog electrical signal, which is conveyed along a conductor that extends from the measurement electrode to the proximal portion of the continuous analyte sensor(s) 202 that is coupled to the sensor electronics module 204. After the continuous analyte monitoring system 104 has been applied to epidermis of the patient, continuous analyte sensor(s) 202 penetrates the epidermis, and the distal portion extends into the dermis and / or subcutaneous tissue under epidermis.

[0075] In certain embodiments, in addition to the planar and coaxial sensors described herein, other configurations of the planar and coaxial continuous analyte sensor(s) 202 may also be used, such as a multi-analyte sensor that includes multiple measurement electrodes, each generating an analog electrical signal that represents the concentration levels of a particular analyte.

[0076] Generally, a single-analyte sensor generates an analog electrical signal that is proportional to the concentration level of a particular analyte. Similarly, each multi-analyte sensor generates multiple analog electrical signals, and each analog electrical signal is proportional to the concentration level of a particular analyte. As an illustrative example, continuous analyte sensor 202 includes a single-analyte sensor configured to measure glucose concentration levels, and another single-analyte sensor configured to measure lactate concentration levels of the patient. As another illustrative example, continuous analyte sensor(s) 202 includes a single-analyte sensor configured to measure glucose concentration levels, and one or more single and / or multi-analyte sensors configured to measure one or more other analyte concentration levels, or analyte ion concentration levels, etc. As yet another illustrative example, continuous analyte sensor(s) 202 includes a multi-analyte sensor configured to measure glucose concentration levels and one or more other analyte concentration levels or analyte ion concentration levels. As yet another illustrative example, continuous analyte sensor(s) 202 includes a multi-analyte sensor configured to measure lactate concentration levels and one or more other analyte concentration levels or analyte ion concentration levels.

[0077] Accordingly, continuous analyte sensor(s) 202 is configured to generate at least one analog electrical signal that is proportional to the concentration level of a particular analyte, andAttorney Docket No.: 0917-PCT01sensor electronics module 204 is configured to convert the analog electrical signal into an analyte sensor count values, calibrate the analyte sensor count values based on the sensitivity profile of the continuous analyte sensor(s) 202 to generate measured analyte concentration levels, and transmit the measured analyte concentration level data, including the measured analyte concentration levels, to a display device, such as display devices 210, 220, 230, and / or 240, via a wireless connection. For example, sensor electronics module 204 can be configured to sample the analog electrical signal at a particular sampling period (or rate), such as every 1 second (1 Hz), 5 seconds, 10 seconds, 30 seconds, 1 minute, 3 minutes, 5 minutes, etc., and to transmit the measured analyte concentration data to the display device at a particular transmission period (or rate), which can be the same as (or longer than) the sampling period, such as every 1 minute (0.016 Hz), 5 minutes, 10 minutes, 30 minutes, at the conclusion of the wear period, etc. Depending on the sampling and transmission periods, the measured analyte concentration data transmitted to the display device include at least one measured analyte concentration level having an associated time tag, sequence number, etc.

[0078] In certain embodiments, continuous analyte sensor(s) 202 can incorporate a thermocouple within, or alongside, the percutaneous wire to provide an analog temperature signal to the sensor electronics module 204, which can be used to correct the analog electrical signal or the measured analyte data for temperature. In other embodiments, the thermocouple can be incorporated into the sensor electronics module 204 above the adhesive pad, or, alternatively, the thermocouple can contact the epidermis of the patient through openings in the adhesive pad.

[0079] In certain embodiments, the sensor electronics module 204 includes, inter alia, processor 233, storage element or memory 234, wireless transmitter / receiver (transceiver) 236, one or more antennas coupled to wireless transceiver 236, analog electrical signal processing circuitry, analog to-digital (A / D) signal processing circuitry, digital signal processing circuitry, a power source for continuous analyte sensor(s) 202 (such as a potentiostat), etc.

[0080] Processor 233 can be a general-purpose or application-specific microprocessor, an application- specific integrated circuit (ASIC), a field programmable gate array (FPGA), etc., that executes instructions to perform control, computation, input / output, etc. functions for the sensor electronics module 204. Processor 233 can include a single integrated circuit, such as a micro processing device, or multiple integrated circuit devices and / or circuit boards working inAttorney Docket No.: 0917-PCT01cooperation to accomplish the appropriate functionality. Tn certain embodiments, processor 233, memory 234, wireless transceiver 236, the A / D signal processing circuitry, and the digital signal processing circuitry can be combined into a system-on-chip (SoC).

[0081] Generally, processor 233 can be configured to sample the analog electrical signal using the A / D signal processing circuitry at regular intervals (such as the sampling period) to generate analyte sensor count values based on the analog electrical signals produced by the continuous analyte sensor(s) 202, calibrate the analyte sensor count values based on the sensitivity profile of the continuous analyte sensor(s) 202 to generate measured analyte concentration levels, and generate measured analyte data from the measured analyte concentration levels, generate sensor data packages that include, inter alia, the measured analyte concentration level data. Processor 233 can store the measured analyte concentration level data in memory 234, and generate the sensor data packages at regular intervals (such as the transmission period) for transmission by wireless transceiver 236 to a display device, such as display devices 210, 220, 230, and / or 240. Processor 233 can also add additional data to the sensor data packages, such as supplemental sensor information that includes a sensor identifier, a sensor status, temperatures that correspond to the measured analyte data, etc. The sensor data packages are then wirelessly transmitted over a wireless connection to the display device. In certain embodiments, the wireless connection is a Bluetooth or Bluetooth Low Energy (BLE) connection. In such embodiments, the sensor data packages are transmitted in the form of Bluetooth or BLE data packets to the display device

[0082] In various embodiments, memory 234 can include volatile and nonvolatile medium. For example, memory 234 can include combinations of random access memory (RAM), dynamic RAM (DRAM), static RAM (SRAM), read only memory (ROM), flash memory, cache memory, and / or any other type of non-transitory computer-readable medium. Memory 234 can store one or more analyte sensor system applications, modules, instruction sets, etc. for execution by processor 233, such as instructions to generate measured analyte data from the analyte sensor count values, etc.

[0083] Memory 234 can also store certain sensor operating parameters 235, such as a calibration slope (or calibration sensitivity), a calibration baseline, etc. In particular, the calibration sensitivity, calibration baseline, and other information related to the sensitivity profile for the sensor electronics module 204 can be programmed into the sensor electronics module 204Attorney Docket No.: 0917-PCT01during the manufacturing process, and then used to convert the analyte sensor electrical signals into measured analyte concentration levels. For example, as discussed above, the calibration slope can be used to predict an initial in vivo sensitivity (Mo) and a final in vivo sensitivity (Mf), which are stored in memory 234 and used to convert the analyte sensor electrical signals into measured analyte concentration levels. In certain embodiments, calibration sensitivity (Mcc) 246 and / or calibration baseline 247 can be stored in memory 234.

[0084] In certain embodiments, sensor electronics module 204 includes electronic circuitry associated with measuring and processing the continuous analyte sensor data, including prospective algorithms associated with processing and calibration of the sensor data. Sensor electronics module 204 can be physically connected to continuous analyte sensor(s) 202 and can be integral with (non-releasably attached to) to continuous analyte sensor(s) 202. Alternatively, sensor electronics module 204 can be physically connected to continuous analyte sensor(s) 202 and releasably attachable to continuous analyte sensor(s) 202. Sensor electronics module 204 can include hardware, firmware, and / or software that enable measurement of levels of analyte(s) via continuous analyte sensor(s) 202. For example, sensor electronics module 204 can include a potentiostat, a power source for providing power to the sensor, other components useful for signal processing and data storage, and a telemetry module for transmitting data from the sensor electronics module to, e.g., one or more display devices. Electronics can be affixed to a printed circuit board (PCB), or the like, and can take a variety of forms. For example, the electronics can take the form of an integrated circuit (IC), such as an Application-Specific Integrated Circuit (ASIC), a microcontroller, and / or a processor.

[0085] Display devices 210, 220, 230, and / or 240 are configured for displaying displayable sensor data, including analyte data, which can be transmitted by sensor electronics module 204. Each of display devices 210, 220, 230. or 240 can include a display such as a touchscreen display 212, 222, 232, and / or 242 for displaying sensor data to a patient and / or for receiving inputs from the patient. For example, a graphical user interface (GUI) can be presented to the patient for such purposes. In certain embodiments, the display devices can include other types of user interfaces such as a voice user interface instead of, or in addition to, a touchscreen display for communicating sensor data to the patient of the display device and / or for receiving patient inputs. Display devices 210, 220, 230, and 240 can be examples of display device 107 illustrated in FIG. 1 used to display sensor data to a patient of the system of FIG. 1 and / or to receive input from the patient.Attorney Docket No.: 0917-PCT01

[0086] In certain embodiments, one, some, or all of the display devices are configured to display or otherwise communicate (e.g., verbalize) the sensor data as it is communicated from the sensor electronics module (e.g.. in a customized data package that is transmitted to display devices based on their respective preferences), without any additional prospective processing required for calibration and real-time display of the sensor data.

[0087] The plurality of display devices can include a custom display device specially designed for displaying certain types of displayable sensor data associated with analyte data received from sensor electronics module. In certain embodiments, the plurality of display devices can be configured for providing alerts / alarms based on the displayable sensor data. Display device 210 is an example of such a custom device. In certain embodiments, one of the plurality of display devices is a smartphone, such as display device 220 which represents a mobile phone, using a commercially available operating system (OS), and configured to display a graphical representation of the continuous sensor data (e.g., including current and historic data). Other display devices can include other hand-held devices, such as display device 230 which represents a tablet, display device 240 which represents a small watch or fitness tracker, medical device 208 (e.g., an insulin delivery device), and / or a desktop or laptop computer (not shown).

[0088] Because different display devices provide different user interfaces, content of the data packages (e.g., amount, format, and / or type of data to be displayed, alarms, and the like) can be customized (e.g., programmed differently by the manufacture and / or by an end patient) for each particular display device. Accordingly, in certain embodiments, a plurality of different display devices can be in direct wireless communication with a sensor electronics module (e.g., such as an on-skin sensor electronics module 204 that is physically connected to continuous analyte sensor(s) 202) during a sensor session to enable a plurality of different types and / or levels of display and / or functionality associated with the displayable sensor data.

[0089] As mentioned, sensor electronics module 204 can be in communication with a medical device 208. Medical device 208 can be a passive device in some example embodiments of the disclosure. For example, medical device 208 can be an insulin pump for administering insulin to a patient. For a variety of reasons, it can be desirable for such an insulin pump to receive and track glucose values transmitted from continuous analyte monitoring systems 104, where continuous analyte sensor 202 is configured to measure at least glucose.Attorney Docket No.: 0917-PCT01

[0090] Further, as mentioned, sensor electronics module 204 can also be in communication with other non-analyte sensors 206. Non-analyte sensors 206 can include, but are not limited to, an insulin pump sensor, stretch sensor, body sound sensor, acoustic gastography sensor, an electrocardiogram (ECG) sensor, a heart rate monitor, a blood pressure sensor, a respiratory sensor, a thermometer, a digital weight scale, sensors or devices provided by display device 107, etc. Non-analyte sensors 206 can also include monitors such as heart rate monitors, blood pressure monitors, pulse oximeters, caloric intake monitors, indirect calorimetry devices, and medicament delivery devices. One or more of these non-analyte sensors 206 can provide data to therapy management engine 114 described further below. In some aspects, a patient can manually provide some of the data for processing by training server system 140 and / or therapy management engine 114 of FIG.1.

[0091] In certain embodiments, non-analyte sensors 206 can further include sensors for measuring skin temperature, core temperature, sweat rate, and / or sweat composition.

[0092] In certain embodiments, the non-analyte sensors 206 can be combined in any other configuration, such as, for example, combined with one or more continuous analyte sensors 202. As an illustrative example, a non-analyte sensor, e.g., a body sound sensor, can be combined with a continuous glucose sensor 202 to form a glucose / body sound sensor used to transmit sensor data to the sensor electronics module 204 using common communication circuitry.

[0093] In certain embodiments, a wireless access point (WAP) can be used to couple one or more of continuous analyte monitoring system 104, the plurality of display devices, medical device(s) 208, and / or non-analyte sensor(s) 206 to one another. For example, the WAP can provide Wi-Fi, Bluetooth and / or cellular connectivity among these devices. Near Field Communication (NFC) and or Bluetooth can also be used among devices depicted in diagram 200 of FIG.2.

[0094] FIG. 3 illustrates example inputs and example metrics that are calculated based on the inputs for use by the therapy management system of FIG. 1, according to some embodiments disclosed herein. In particular, FIG.3 provides a more detailed illustration of example inputs and example metrics introduced in FIG. 1.

[0095] FIG. 3 illustrates example inputs 130 on the left, application 106 and DAM 116 in the middle, and metrics 132 on the right. In certain embodiments, each one of metrics 132 can correspond to one or more values, e.g., discrete numerical values, ranges, or qualitative valuesAttorney Docket No.: 0917-PCT01(high / medium / low, stable / unstable, etc.). Application 106 obtains inputs 130 through one or more channels (e.g., manual patient input, sensors, other applications executing on display device 107, an EMR system, etc.). As mentioned previously, in certain embodiments, inputs 130 can be processed by DAM 116 to output a plurality of metrics, such as metrics 132. Inputs 130 and metrics 132 can be used by training server system 140 and therapy management engine 114 to both train and deploy one or more machine learning models for determining a specific diabetic state classification, providing various treatment recommendations to manage the patient’s health based on the patient’s classification, monitoring the progression and / or regression of the patient’s diabetic state over time, and other functionalities described herein.

[0096] In certain embodiments, starting with inputs 130, patient statistics, such as one or more of age, gender, height, weight, BMI, body composition (e.g., % body fat), stature, build, or other information can also be provided as an input. In certain embodiments, patient statistics are provided through a user interface, by interfacing with an electronic source such as an electronic medical record, and / or from measurement devices. In certain embodiments, the measurement devices include one or more of a wireless, e.g., Bluetooth-enabled, weight scale and / or camera, which can, for example, communicate with the display device 107 to provide patient data. Among patient statistics, for example, age can affect a patient’s glucose metabolism, insulin sensitivity, and metabolic fitness. Older patients (e.g., greater than 70 years old) generally can be expected to have slower rate of change of glucose levels to return to baseline glucose levels and less glucose variability when compared to younger patients (e.g., less than 40 years old). For example, for a younger patient, a maximum glucose level can be reached within 30 minutes, but for older patients, it can take 45 minutes to reach a greater maximum glucose level than the younger patient in response to the same or similar meal. Additionally, an area under the curve of the patient’s glucose trace can be greater for older patients, as older patients can take longer (e.g., 120 minutes) to return to a glucose baseline level following a meal when compared to a younger patient consuming the same meal whose return to baseline glucose levels can occur within 90 minutes of the meal.

[0097] In certain embodiments, treatment / medication information is also provided as an input. Medication information can include information about the type, dosage, and / or timing of when one or more medications are to be taken by the patient. Incorrect timing or dosage of medications, such as insulin, can lead to glucose level fluctuations as well as changes in various other glucose metrics. Treatment information can include information regarding different lifestyle habitsAttorney Docket No.: 0917-PCT01recommended by the patient’s physician. For example, the patient’s physician can recommend a patient follow specific diet recommendations (e.g., types of calories consumed), exercise at a specific time during the day for a specific duration, eat a meal at certain days and / or times, or cut calories by 500 to 1,000 calories daily to improve analyte levels (e.g., glucose levels, for example) and therefore improve their diabetic state classification and / or reduce the risk of developing diabetes. In certain embodiments, treatment / medication information can be provided through manual patient input.

[0098] In certain embodiments, analyte sensor data can also be provided as input, for example, through continuous analyte monitoring system 104. In certain embodiments, analyte sensor data can include glucose levels measured by at least a single analyte sensor (or multi-analyte sensor) in continuous analyte monitoring system 104.

[0099] In certain embodiments, input can also be received from one or more non-analyte sensors, such as non-analyte sensors 206 described with respect to FIG. 2. Input from such non-analyte sensors 206 can include information related to a heart rate, a respiration rate, oxygen saturation, blood pressure, or a body temperature (e.g. to detect illness, physical activity, etc.) of a patient and / or measurements of variations, averages, derivatives, or any other multi-measurement analytical calculations between at least two points of non-analyte and / or analyte data. In certain embodiments, electromagnetic sensors can also detect low-power radio frequency (RF) fields emitted from objects or tools touching or near the object, which can provide information about patient activity or location.

[0100] In certain embodiments, food consumption information is also provided as input. Food consumption information can include information about one or more of meals, snacks, and / or beverages, such as one or more of the size, nutritional composition (carbohydrate, fat, protein, etc.), sequence of consumption, and time of consumption. In certain embodiments, food consumption can be provided by a patient through manual entry, by providing a photograph through an application that is configured to recognize food types and quantities, and / or by scanning a bar code or menu. In various examples, meal size can be manually entered as one or more of calories, quantity (“three cookies”), menu items (“Royale with Cheese”), and / or food exchanges (1 fruit, 1 dairy). In some examples, meal information can be received via a convenient user interface provided by application 106.Attorney Docket No.: 0917-PCT01

[0101] In certain embodiments, food consumption information (the type of food (e.g., liquid or solid, snack or meal, etc.) and / or the composition of the food (e.g., carbohydrate, fat, protein, etc.)) can be determined automatically based on information provided by one or more sensors. Some example sensors can include body sound sensors (e.g., abdominal sounds can be used to detect the types of meal, e.g., liquid / solid food, snack / meal, etc.), radio-frequency sensors, cameras, hyperspectral cameras, and / or glucose sensors to determine the type and / or composition of the food. In certain embodiments, high carbohydrate meals can cause a more immediate postprandial glucose level spike, while fiber and protein, either alone or in combination with carbohydrates, can moderate glucose absorption. Therefore, meal composition information can cause therapy management engine 114 to expect various glucose metrics in response to the specific composition of meal the patient consumed.

[0102] In certain embodiments, therapy management engine 114 can recommend the patient consume a specific meal kit or prepared meal (e.g., a prepared meal or series of prepared meals with known ingredients and nutrition or an at-home meal preparation kit). The nutrition of the meal kit can be manually input by the patient or the meal can be packaged with a QR code or other unique identifier which can be scanned to upload the nutrition information to therapy management engine 114. In certain embodiments, the patient can submit a photo of the meal including the time of the meal and a general description of the meal. Therapy management engine 114 can estimate the nutritional information of the meal based on the patient submitted photo. Based on the known or estimated nutritional information, therapy management engine 114 can correlate specific glucose metrics to various types of meals and provide specific recommendations based on meals that caused a healthy glucose response or meals that caused an unhealthy glucose response (e.g., “white rice is not recommended as it causes a glucose level spike.” or “balance your meal with equal parts protein, fat, and carbohydrate”).

[0103] In certain embodiments, medical history and / or disease diagnoses (e.g., diabetes, liver disease, kidney disease, heart disease, obesity, etc.) can be provided as an input. For example, the patient can have an existing diabetic state classification and / or diagnosis of diabetes and this diagnosis can be provided through manual patient input. In certain embodiments, disease diagnoses can also be provided by interfacing with an electronic source such as an electronic medical record. In certain embodiments, input related to liver disease can include a metric of fat content of the patient’s liver. High hepatic fat content can be associated with insulin resistance,Attorney Docket No.: 0917-PCT01which can cause higher glucose levels and less glucose response to insulin. For example, increasing fat content in the liver can demonstrate pre-diabetes and eventually lead to the development T2D. Patients with a liver fat content of 3.8% can demonstrate a similar maximum glucose level to historical healthy patients, but can experience a delay in return to baseline glucose levels by 30 minutes to 90 minutes. Patients with a higher liver fat content, such as 5.5%, the delay in return to baseline glucose levels can be delayed by 90 minutes to 2.5 hours and the maximum glucose level can exceed the maximum glucose level of historical healthy patients. Therefore, a patient who does not have a disease diagnosis of diabetes but known to have a high liver fat content, such as 5.5%, can experience a higher maximum glucose level and a slower return to baseline glucose levels when compared to historical healthy patients who consumed the same meal.

[0104] Additionally, kidney disease can cause impaired gluconeogenesis which can decrease glucose levels and delayed insulin clearance which can further decrease glucose levels. Kidney disease can also affect glucose clearance in urine and / or impaired reabsorption of glucose which can cause an increase or decrease in the patient’s glucose levels.

[0105] In certain embodiments, exercise information is also provided as an input. Exercise information can be any information surrounding activities requiring physical exertion by the patient. For example, exercise information can range from information related to low intensity (e.g., walking a few steps) and high intensity (e.g., five mile run) physical exertion. In certain embodiments, the exercise information can comprise information related to HIIT, resistance training, or Zone 2 training. In certain embodiments, exercise information can also be provided through manual patient input suggesting the patient will begin a specific exercise type and / or with certain exercise parameters.

[0106] In certain embodiments, exercise information can be provided or determined based on information provided, for example, by non-analyte sensors 206 (e.g., an insulin pump sensor, a temperature sensor, a heart rate monitor, a wearable blood pressure monitor, an accelerometer sensor on a wearable device such as a watch, fitness tracker, and / or patch, etc.). In certain embodiments, exercise information can be provided or determined based on information provided, for example, by continuous analyte monitoring system 104 (e.g., it can be deduced that the host engaged in exercise based on their glucose data). The exercise information provided by analyteAttorney Docket No.: 0917-PCT01and non-analyte sensors can be used as input into a model trained for predicting whether the patient is engaging in exercise and / or predicting the types and / or intensity of such exercise. Moderate exercise can lead to lower blood glucose levels, while high intensity exercise can lead to increased blood glucose levels due to a patient’s adrenaline response.

[0107] In certain embodiments, time can also be provided as an input, such as time of day or time from a real-time clock. For example, in certain embodiments, input analyte data can be timestamped to indicate a date and time when the analyte measurement was taken for the patient.

[0108] Patient input of any of the above-mentioned inputs 130 can be provided through continuous analyte monitoring system 104, non-analyte sensors 206, and / or a user interface, such a user interface of display device 107 of FIG. 1. As described above, in certain embodiments, DAM 116 determines or computes the patient’s metrics 132 based on inputs 130. An example list of metrics 132 is shown in FIG. 3.

[0109] In certain embodiments, glucose metrics can be determined from sensor data (e.g., glucose measurements obtained from a continuous glucose sensor of continuous analyte monitoring system 104). For example, glucose metrics refer to time-stamped glucose measurements or values that are continuously generated and stored over time. In some examples, glucose metrics can also be determined, for example, based upon historical data in particular situations, e.g., given a combination of food consumption, insulin, and / or exercise. In certain embodiments, a dawn effect pattern can be determined from glucose metrics. A dawn effect pattern can refer to an increase in a patient’s glucose level in the morning, as the patient wakes from sleep. For example, a dawn effect can be determined when the patient’s glucose level spikes to a level of more than 120 mg / dL without cause (e.g.. not in response to consumption of a meal or an exercise session, for example), which can affect insulin sensitivity and glucagon production. Further, a dawn effect pattern can be based on the time of day the glucose level spike occurs. For example, therapy management engine 114 can determine the time of day of the glucose spike to determine whether the glucose spike without cause can be attributed to the dawn effect.

[0110] In certain embodiments, a minimum and maximum glucose level can be determined from sensor data. For example, a daily minimum and maximum glucose values for each day over a specified amount of time (e.g., a week or a month) can be determined. In certain embodiments, the minimum and maximum glucose levels can be determined based on an average minimum andAttorney Docket No.: 0917-PCT01maximum over a specified amount of time (e.g., a week or a month). In certain embodiments, DAM 116 can continuously or periodically calculate a normal glucose range and time-stamp and store the corresponding information in the patient’s profile 118.

[0111] In other embodiments, a normal minimum and maximum glucose level can be determined from population data (e.g., from data records or historical patients with various diabetic state classifications). In such embodiments, each patient can have personalized, customized, acceptable glucose minimum and / or maximum glucose values, which can be determined based on time periods when the patient is in a fasting state or during a meal, for example.

[0112] In certain embodiments, a glucose baseline can be determined from sensor data (e.g., glucose measurements obtained from a continuous glucose sensor of continuous analyte monitoring system 104). A glucose baseline represents a patient’s normal glucose levels during periods where fluctuations in glucose production is typically not expected. A patient’s baseline glucose level is generally expected to remain constant over time, unless challenged through an action such as consuming food or exercise by the patient, for example. The glucose baseline level can refer to an average glucose baseline over a period of time such as a day, a week, a month, a year, or more than a year. For example, a daily glucose baseline can fluctuate from day to day to reflect one or more external conditions of the user such as stress, lack of sleep, a prior meal consumed, a physical activity, trauma, illness, and / or other events.

[0113] Additionally, a patient’s baseline glucose level can also change based on the patient’s health, specifically an improvement or decline in kidney and / or liver health. Further, each patient can have a different glucose baseline. In certain embodiments, a patient’s glucose baseline can be determined by calculating an average of glucose levels over a specified amount of time where fluctuations are not expected.

[0114] For example, the baseline glucose level for a patient can be determined over a period of time when the patient is sleeping, sitting in a chair, or other periods of time where the patient is sedentary and not consuming food or medication which would reduce or increase glucose levels. In certain embodiments, DAM 116 can continuously, semi-continuously, or periodically calculate a glucose baseline and time-stamp and store the corresponding information in the patient’s profile 118. In certain embodiments, DAM 116 can calculate the glucose baseline using glucose levelsAttorney Docket No.: 0917-PCT01measured over a period of time where the patient is sedentary, the patient is not consuming glucose-heavy foods, and where no external conditions exist that would affect the glucose baseline.

[0115] In certain embodiments, a glucose baseline includes a baseline value for any of the glucose metrics described herein. For example, a glucose baseline can be determined for a glucose rate of change, a glucose minimum, a glucose maximum, a glucose variability, a glucose area under the curve, etc. The glucose baseline can be determined over a period of time such as a day, a week, a month, a year, or more than a year. In certain embodiments, therapy management engine 114 can determine a change from average glucose levels or patterns (e.g., based on historical patient glucose baseline measurements over time) and identify the glucose measurement or series of glucose measurements as an outlier to be excluded from glucose baseline calculations. For example, a change from average glucose levels or patterns relative to the patient’s glucose baseline can be expected and excluded during certain time periods where the patient usually deviates from normal eating habits, such as during the holidays (e.g., holidays where the patient can consume traditional foods high in fat or other specific nutrients, or during Thanksgiving or Christmas).

[0116] In certain other embodiments, DAM 116 can use glucose levels measured over a period of time where the patient is, at least for a subset of the period of time, engaging in exercise and / or consuming glucose and / or an external condition exists that would affect the glucose baseline level. In this case, in some examples, DAM 116 can first identify which measured glucose values are to be used for calculating the baseline glucose level by identifying glucose values that can have been affected by an external event, such the consumption of food, exercise, medication, or other perturbation that would disrupt the capture of a glucose baseline measurement. DAM 116 can then exclude such measurements when calculating the glucose baseline level of the patient. In some other examples, DAM 116 can calculate the glucose baseline level by first determining a percentage of the number of glucose values measured during a specific time period that represent the lowest glucose values measured. DAM 116 can then take an average of this percentage to determine the glucose baseline level.

[0117] In certain embodiments, a glucose rate of change can be determined from glucose levels (e.g., glucose measurements obtained from a continuous glucose sensor of continuous analyte monitoring system 104). A glucose rate of change refers to a rate that indicates how one or more time-stamped glucose measurements or values change in relation to one or more other time-Attorney Docket No.: 0917-PCT01stamped glucose measurements or values. A glucose rate of change can be monitored following the patient consuming a meal, when the patient experiences an increase in glucose. Therapy management engine 114 can determine the rate of change of glucose as the patient’s glucose levels return to baseline following consumption of a meal. Glucose rates of change can be determined over one or more seconds, minutes, hours, days, etc. Further, glucose rate of change can be positive, negative, or an absolute value. In certain embodiments, a glucose rate of change above or below a threshold can be determined. A rapid rise in glucose outside of a threshold, not related to exercise or a meal, can be indicative of organ dysfunction (e.g., kidney and / or liver dysfunction).

[0118] In certain embodiments, a glucose variability can be determined from the analyte data. In some examples, the glucose variability can be determined based on the variability of glucose levels (e.g., oscillations in the patient’s glucose levels over the course of the day) as compared to an acceptable glucose variability range (e.g., between 26-28 mg / dL from baseline glucose levels). In certain embodiments, a time-in-range metric (not shown) can be determined from the glucose data. For example, with an established upper limit and lower limit, the time period during which the glucose data is between the upper and lower limits can be determined. The time-in-range may be determined for individual instances of the glucose data being in range, or may be determined over a predetermined length of time (e.g., one day) for which each of the individual in range periods are summed.

[0119] In certain embodiments, insulin sensitivity and / or insulin deficiency can be determined using historical data, real-time data, or a combination thereof, and can, for example, be based upon one or more inputs 130, such as one or more of food consumption information, continuous analyte sensor data, non-analyte sensor data (e.g., insulin delivery information from an insulin device), etc. Insulin sensitivity refers to how responsive a patient’s cells are to insulin and can assist in diagnosing insulin resistant pre-diabetes or insulin resistant T2D. Improving insulin sensitivity for a patient can help to reduce insulin resistance in the host. Insulin deficiency refers to the rate at which the patient’s cells produce insulin. Insulin deficiency can lead to decreased glucose uptake by cells within the patient’s body, therefore increasing the patient’s blood glucose levels.

[0120] In certain embodiments, insulin on board can be determined using non-analyte sensor data input (e.g., insulin delivery information) and / or known or learned (e.g. from patient data) insulin time action profiles, which can account for both basal metabolic rate (e.g., update of insulinAttorney Docket No.: 0917-PCT01to maintain operation of the body) and insulin usage driven by activity or food consumption. Nonanalyte sensor data input can further allow for the identification of insulin pump infusion set issues (e.g., pump occlusions) which can lead to higher glucose levels due to incorrect insulin dosing. Additionally, non-analyte sensor data can include information on the location of an insulin pump, for example. If therapy management engine 114 determines the patient’s insulin pump is repeatedly using the same injection site, the patient can experience scarring which can affect the absorption of insulin and, therefore, cause variability in the patient’s blood glucose levels.

[0121] In certain embodiments, health and sickness metrics can be determined, for example, based on one or more of patient input (e.g., pregnancy information or known sickness information), from physiologic sensors (e.g., temperature), activity sensors, or a combination thereof. In certain embodiments, based on the values of the health and sickness metrics, for example, a patient’s state can be defined as being one or more of healthy, ill, rested, or exhausted. Illness and stress, for example, can cause a rise in blood glucose levels due to the patient’s production of hormones such as cortisol and adrenaline. Increases in hormones when the patient is sick or stressed can cause higher glucose levels and / or muted glucose variability. Further, if the patient is exhausted and / or the patient has been suffering from poor sleep quality, the patient can experience higher, more variable blood glucose levels, increased insulin resistance, and / or changes in appetite and / or weight.

[0122] In certain embodiments, disease stage metrics, such as for kidney and / or liver disease, can be determined, for example, based on one or more of patient input or output provided by therapy management engine 114 illustrated in FIG. 1. In certain embodiments, example disease stages for liver disease can include an inflammation stage (e.g., early stage where the patient’s liver is enlarged or inflamed), a fibrosis stage (e.g., stage with signs of scar tissue in the inflamed liver), a cirrhosis stage (e.g., stage with signs of severe scar tissue in the inflamed liver), and an end-stage liver disease (ESLD). In certain embodiments, example disease stages for kidney disease can include various stages of kidney disease (e.g.. Stage 1 - Stage 5) and / or chronic kidney disease. Additionally, example disease stages can include information on heart disease and / or obesity.

[0123] In certain embodiments, the meal state metric can indicate the state the patient is in with respect to food consumption. For example, the meal state can indicate whether the patient isAttorney Docket No.: 0917-PCT01in one of a fasting state, pre-meal state, eating state, post-meal response state, or stable state. In certain embodiments, the meal state can also indicate nourishment on board, e.g., meals, snacks, or beverages consumed, and can be determined, for example from food consumption information, time of meal information, and / or digestive rate information, which can be correlated to food type, quantity, and / or sequence (e.g., which food / beverage was eaten first.).

[0124] In certain embodiments, meal habits metrics are based on the content and the timing of a patient’s meals. For example, if a meal habit metric is on a scale of 0 to 1, the better / healthier the meal consumed by the patient, the higher the meal habit metric of the patient will be to 1, in an example. Better and / or healthier meals can be defined as those that do not drive glucose levels of a patient out of a normal range for the patient (e.g., 70-180 mg / dL glucose or the patient’s desired range). Also, the more the patient’s food consumption adheres to a certain time schedule, the closer their meal habit metric will be to 1, in the example. In certain embodiments, the meal habit metrics can reflect the contents of a patient’s meals where, e.g., three numbers can indicate the percentages of carbohydrates, proteins and fats.

[0125] In certain embodiments, medication habit metrics are based on the patient’s prescribed medications and a determination of whether the prescribed medications can have an effect on the patient’s glucose levels. For example, by analyzing a patient’s medication habits, DAM 116 determines whether the patient’s medications may impact the patient’s glucose measurements at a particular time. Based on the patient’s medication habits, DAM 116 can determine whether the patient’s glucose levels are a result of medication consumption or worsening diabetic state classification, for example. Medication habit metrics can be time-stamped so that they can be correlated with the patient’s glucose levels at the same time.

[0126] In certain embodiments, medication adherence is measured by one or more metrics that are indicative of how committed the patient is towards their medication regimen. In certain embodiments, medication adherence metrics are calculated based on one or more of the timing of when the patient takes medication (e.g., whether the patient is on time or on schedule), the type of medication (e.g., is the patient taking the right type of medication), and the dosage of the medication (e.g., is the patient taking the right dosage).

[0127] In certain embodiments, body temperature metrics can be calculated by DAM 116 based on inputs 130, and more specifically, non-analyte sensor data from a temperature sensor. InAttorney Docket No.: 0917-PCT01certain embodiments, heart rate metrics can be calculated by DAM 116 based on inputs 130, and more specifically, non-analyte sensor data from a heart rate sensor. In certain embodiments, respiratory rate metrics can be calculated by DAM 116 based on inputs 130. and more specifically, non-analyte sensor data from a respiratory rate sensor.Example Methods for Predicting a Patient’s Diabetic State Classification Using Continuously Monitored Glucose Levels

[0128] FIG. 4 is a flow diagram illustrating an example method 400 for execution by therapy management engine 114 to classify a diabetic state of a patient, provide personalized therapy management guidance for managing the patient’s health based on the patient’s diabetic state, and monitor the progression and / or regression of the patient’s diabetic state over time. The classification and / or monitoring of the patient’s diabetic state is provided by therapy management engine 114 based on the continuous analyte data of the patient. For example, glucose levels generated by a continuous analyte monitoring system 104 is used to classify the patient, and to monitor and update the classification, and / or assess the patient’s diabetic state over time. In addition, other analyte data such as lactate data may also be used. Additionally, patient information, and non-analyte sensor data mentioned above may be used during the performance of method 400. Method 400 is described below with reference to FIGs. 1 and 2 and their components.

[0129] As discussed above, therapy management engine 114 may use one of a variety of algorithms or models to classify the diabetic state of the patient, provide personalized recommendations for managing the patient’s health based on the patient’s diabetic state, and / or monitor the progression and / or regression of the patient’s diabetic state over time. As described above, the inputs to these models may include glucose data or other analyte data (e.g., received by continuous analyte monitoring system 104), non-analyte data, and / or other patient information (e.g., retrieved from the patient’s profile or received via patient inputs).

[0130] For example, in embodiments where a rules-based model is used, the therapy management engine 114 can execute a rule to determine a patient’s diabetic state based on the patient’s continuous analyte data. For example, a patient may be determined to have a severe T2D classification if the patient’s glucose data is consistent with historical patients having a severe T2D classification.Attorney Docket No.: 0917-PCT01In certain embodiments, the rales can become more granular, such that a combination of rules and / or inputs would allow therapy management engine 114 to output an even more accurate classification of a patient’s diabetic state. For example, a primary metric from the list of glucose metrics provided above may provide a first level of differentiation, and in these cases, a secondary metric (or further grouping of metrics, in series or parallel) from the list described above is used to further classify these patients. As an example, in insulin deficient pre-diabetes, after consuming a small amount of carbohydrate (e.g., 20 g), glucose levels are likely to rise to levels higher than a healthy individual, similar to someone with insulin resistant pre-diabetes. Therefore, the patterns in how the glucose level rises or returns to normal may provide a first level of classification, but may not provide a second level (e.g., root cause) classification. Therefore, an additional metric can be used to classify these patients. In one example, the metric is time to return back to baseline. For example, a patient with insulin deficient prediabetes glucose levels will likely return back to a baseline faster than (e.g., a higher rate of change, or a shorter time period) someone with insulin resistant diabetes but slower (e.g., a lower rate of change or a longer time period) than a healthy individual. A rale can be implemented by therapy management engine 114 to determine the patient has insulin deficient pre-diabetes if an average baseline glucose level of the patient, calculated over a period of time, has a value that is less than a predetermined difference from a predetermined amount (e.g., having at most a 5mg / dL difference from a value of 105mg / dL), and / or a pattern of the patient’s glucose indicates a return to baseline glucose values by two hours following a glucose excursion or a glucose variating event (e.g., a meal). Another example rule can be implemented by therapy management engine 114 to determine that the patient has insulin resistant mild T2D if the patient’s glucose metrics demonstrate a fasting glucose level between 100 mg / dL and 150 mg / dL, a low amount of glucose variability at baseline when compared to a patient without diabetes, and a post-prandial glucose spike between 150 mg / dL and 225 mg / dL.

[0131] Another metric to differentiate can be the glucose response to light exercise. For example, in insulin-deficient diabetes, an individual would have a robust glucose response to light exercise, while an individual with insulin-resistant diabetes would have a reduced glucose response or impaired return to baseline due to general insulin resistance. Any of the metrics described above with respect to Figures 1 -3 can be used to provide layered classification of the patient. For example, the system may classify users by postprandial glucose spike, repeated hypoglycemia events, timein-range (TIR). rate of decline or increase of glucose in response to expected or reported meal, andAttorney Docket No.: 0917-PCT01area under the curve after ingestion of a meal. The term meal is used generally to describe any substance being ingested that contains one or a mixture of carbohydrates, protein and / or fat, and can include actual meals or glucose drinks, lactate drinks, ketone drinks, prescribed meals, and other items that may cause a change in glucose levels and / or a change in other analytes.

[0132] In some examples, the rule also takes into account patient inputs, such as patient age, disease diagnoses, and / or additional analyte data (e.g., lactate or ketone data). Drug-specific information, comorbidities, and other healthcare information may be incorporated for classifying the diabetic state of the patient..

[0133] In certain embodiments, instead of or in conjunction with a rules-based model, an AI / ML model can be used to output a classification of a patient’s diabetic state. Some or all of the inputs described above can be used as input into the model. In certain embodiments, the model is trained using a dataset, including historical population-based data of many patients, who have already been determined to have various diabetic state classifications. In such an example, the training dataset is labeled with such determinations.

[0134] In some embodiments, to provide accurate classification of diabetic state to be made for a patient, the therapy management engine 114 can identify periods when the patient’s glucose levels reflect their typical baseline as a comparator to determine glucose patterns and excursions for the patient. Therapy management engine 114 can filter out glucose levels that seem to be impacted by external factors and not consistent with the patient’s typical baseline. External factors that can effect a patient’s glucose levels may include an activity performed by the patient (e.g., consumption of alcohol), a short term illness (e.g., a bacterial or viral infection), an acute event related to a chronic illness (e.g.. a multiple sclerosis flare up), and / or any other event that may cause the patient’s glucose levels to demonstrate abnormal levels and / or patterns. Upon detecting that the patient’s glucose levels have been or are going to be impacted by one of the factors discussed above, therapy management engine 114 can refrain from classifying the patient based on their glucose levels until the events have passed or the events have been resolved to allow the patient’s glucose levels to return to baseline. Alternatively, the glucose levels collected during the events can be excluded from the classification of the patient’s diabetic state using a filter or weight assignment (e.g., less or zero weight assigned to the glucose levels).Attorney Docket No.: 0917-PCT01

[0135] For example, based on the patient’ s inputs and / or data from analyte and / or non-analyte sensors, therapy management engine 114 can monitor for one or more actions that can temporarily alter the patient’s glucose levels, metabolic function, insulin production and / or insulin sensitivity. As an example, therapy management engine 114 can determine the patient has consumed alcohol recently based on patient input reporting alcohol consumption. Based on the detected alcohol consumption, therapy management engine 114 can refrain from classifying the patient until a trigger event to resume classification of the patient’s diabetic state. The trigger event can be a threshold period of time, a determination that the patient’s glucose levels returned to normal levels (e.g., a historical baseline glucose level), or a determination that the condition which lead to the pause has been resolved.

[0136] As another example, therapy management engine 114 can determine, either from patient input, the patient profile, and / or glucose or non-analyte data, whether the patient is suffering from an illness. For example, a short term illness such as a viral or bacterial infection can impact the patient’s glucose levels and, therefore, cause the patient to be misclassified if the patient’s glucose levels were to be used for classification during such illness. In another example, a long term illness with one or more periods of acute flare ups requiring treatment includes multiple sclerosis flare ups requiring steroid treatments, which can also impact the patient’s glucose levels. In such a case, therapy management engine 114 can refrain from classifying the patient until a trigger event.

[0137] Alternatively, following a determination of an external factor affecting glucose levels, therapy management engine 114 can collect glucose values that are used provide a classification of the diabetic state of the patient based on the affected glucose levels. The classification of the diabetic state of the patient may be output in association with the external factor and / or while noting that the classification may be an inaccurate classification. In another example, therapy management engine 114 can correct the patient’s classification based on a known external factor.

[0138] In one example, the classification of the patient is provided after a predetermined amount of inputs either from patient input, the patient profile, and / or glucose or non-analyte data, have been collected such that the data is representative of an average pattern for the patient. For example, by taking into account a predetermined period of time for collecting data (e.g., 12 hours, a full day, 3 days, a week or an entire wear session), metrics can be generated that represent aAttorney Docket No.: 0917-PCT01normalized glucose profile (glucose responses) for the patient and therefore, account for unknown external factors or anomalies that may affect point in time measurements. In another example, a threshold number of values can be collected before metrics are used to classify the patient. The number of values may be time dependent or may be selected to include values within a specific range, such that outliers are discarded.

[0139] Once a classification is performed, in certain embodiments, therapy management engine 114 can continue monitoring and verifying the patient’s classified diabetic state over time. For example, therapy management engine 114 can monitor a patient’s classification of diabetic state over time to verify whether the classification was accurate. In some examples, if the therapy management engine 114 determines that the classification is not accurate or does not have a desired level of certainty, the classification can be updated based on new data until it is determined that the classification is within the desired level of certainty. The level of certainty can be based on the number of outlier glucose values, known or suspected external factors present (e.g., that the glucose values are or likely to be a result of a short term illness, an acute event related to a long term illness, or alcohol consumption, for example), a numerical confidence level calculated, or based on the glucose values being within a certain variability (e.g., within a variability threshold).

[0140] Further, over time, therapy management engine 114 can determine whether the patient’s diabetic state classification has remained the same, or if the patient’s diabetic state classification is progressing towards a new classification. The progression of the patient’s diabetic state can be to a less severe classification (e.g., from prediabetes to healthy, from severe insulin resistant diabetes to mild insulin resistant diabetes), a more severe classification (e.g., mild to severe T2D, prediabetes to diabetes, etc.), or to a compounded classification (e.g., from insulin resistant to insulin resistant and insulin dependent). The classification of the patient’s diabetic state can change over time due to changes in the patient’s organ health (e.g.. liver, kidney or pancreas health), metabolic fitness, body composition, age, overall health, change in medication, or other factors that affect the patient’s ability to regulate their glucose levels.

[0141] Based on the patient’s diabetic state classification over time, therapy management engine 114 can provide specific therapy management guidance to the patient for actions and / or lifestyle changes to improve or maintain the patient’ s diabetic state. Therapy management engine 114 may also provide therapy management guidance to the patient during any acute events, shortAttorney Docket No.: 0917-PCT01term illnesses, and / or alcohol consumptions to limit the effect of such factors on the patient’s diabetic state or the patient’ s overall health. Additionally, the therapy management guidance from therapy management engine 114 can provide positive feedback when the patient’s actions are improving the patient’s diabetic state over time.

[0142] The therapy management engine 114, classifies users within each category based on level of glucose control and disease severity according to the metrics provided above, providing customized decision support recommendations. For example, there may be five categories of diabetes, each with three levels of severity, and different root causes, resulting in different recommendations on duration of wear, sensor configuration, and treatment (e.g., disease management or health and fitness activities) recommendations. Decision support treatment and management recommendations may be provided depending on whether a patient is insulin deficient, insulin resistant, or metabolically unfit, with specific recommendations for exercise, diet, medication, and additional analyte monitoring.

[0143] These recommendations can include a personalized view of nutrition and exercise recommendations and tracking, along with recommended spacing between sensor wear. Recommendations may be adjusted for cultural lifestyle changes, holidays, and other events likely to disrupt normal routines. The system may also provide decision support for users without diabetes to maintain or improve metabolic health, with recommendations for diet, exercise, and sensor wear frequency. The system may incorporate contextual factors such as holidays, travel, and lifestyle disruptions into sensor wear scheduling and recommendations. The system may recommend additional analyte monitoring, such as lactate, potassium, or creatinine, based on user classification and comorbidities.

[0144] Returning to exemplary method 400, method 400 begins at block 402 by monitoring at least glucose concentration levels of a patient during a time period using one or more glucose sensors to obtain glucose data. The period of time can be a single session or multiple sessions during which the patient is wearing a continuous analyte monitoring system (such as continuous analyte monitoring system 104). In certain embodiments, obtaining the glucose data includes determining one or more glucose metrics for the patient based on the glucose concentration levels. The glucose metrics include at least one of a glucose minimum or maximum, glucose rate of change, glucose baseline, glucose variability, and / or other types of glucose metrics as described inAttorney Docket No.: 0917-PCT01reference to FIG.3. As used herein, monitoring at least glucose concentration levels of the patient can include receiving analyte concentration data of the patient, wherein the analyte concentration data includes at least glucose concentration levels, and processing the continuous analyte concentration data to identify analyte metrics of the patient, including at least glucose metrics.

[0145] At block 404, therapy management engine 114 proceeds by determining a classification of a patient’s diabetic state based on at least the glucose data of the patient. As described further in reference to FIG. 5, therapy management engine 114 provides one of the classifications to the patient including a disease type: healthy patient, pre-diabetes, Typel, Type 2 diabetes, gestational diabetes, metabolic dysfunction, or other dysglycemic classification based on the patient’s glucose data. In addition, the classification can include the severity of the disease. For example, a patient may be classified as moderate or severe Type 1 or Type 2, as insulin dependent vs. non-insulin dependent Type 2, as early or late stage diabetes, as gestational vs. post partem diabetes, liver disfunction, etc. In one example, a severity score may be computed based on the metrics and assigned to the patient as part of the diabetic state of the patient. Additionally or alternatively, the classification can include the root cause of the glycemic patterns of a patient. For example, insulin deficiency, insulin resistance, genetics, hormonal issues or fluctuations, metabolic inflexibility, etc., can be examples of root causes for a patient’s glycemic response. Certain metrics can be used to determine the root cause of a patient’s glucose patterns and may be provided as part of the classification of the diabetic state of the patient.

[0146] In certain embodiments, to determine a classification for the patient’s diabetic state, the therapy management engine 114 compares the patient’s glucose data with one or more predetermined thresholds or patterns. The threshold or pre-determined patterns may be based on historical patterns for the patient (e.g., to indicate a progression of disease for a patient or to serve as a baseline), can be based on population information such as historical glucose data of patients with various diabetic state classifications, and / or can be generated according to clinical guidelines and standards of care. The pre-determined thresholds and patterns can be compared against glucose values of the patient (or metrics computed thereof) to determine the diabetic state of the user. For example, if a patient’s metrics (e.g., glucose metrics) are consistent with thresholds or patterns associated with severe T2D, therapy management engine 114 may classify the patient as a patient with severe T2D. Similarly, if a patient’s baseline or historical values show a progression of disease the progression can be used to classify the diabetic state of the patient.Attorney Docket No.: 0917-PCT01

[0147] Following the determination of the classification at block 404, therapy management engine 114 may then optionally proceed to block 406. At block 406, therapy management engine 114 determines a root cause for the determined classification of the diabetic state. For example, therapy management engine 114 utilizes one or more metrics, including insulin sensitivity and / or glucose metrics (e.g., the patient’s baseline glucose fluctuation, the patient’s glucose spike following a meal, and / or the patient’s post- prandial return to baseline glucose levels following a glucose peak), to determine the root cause of the patient’s glycemic patterns. For example, where the patient has glucose spikes or other glucose excursions, has a delayed return to baseline from excursions, or other glucose patterns that indicate a glycemic response that may be problematic for a patient, the system may determine a root cause for such patterns. For example, the root cause may be insulin resistance .insulin deficiency, metabolic inflexibility, or other comorbid conditions.

[0148] In certain embodiments, therapy management engine 114 determines a root cause of a patient’s determined classification only if the patient has a certain type classification for their diabetic state. For example, if the patient’s classification is a healthy patient or a patient with T1D or severe T2D, therapy management engine 114 may not determine the root cause for the patient’s determined classification, whereas if a patient is classified as having mild T2D, pre-diabetes, gestational diabetes, or other dysglycemic issues, the system may classify the root cause of the patient’s glycemic condition. Determined classifications that involve the root cause determination to provide accurate therapy management guidance are described in reference to FIG. 5.Specifically, the determination of the root cause of the patient’s determined classification of block 406 is described in more detail in reference to blocks 518, 528, and 540.

[0149] For example, for a patient whose determined classification is a pre-diabetic patient, a patient with mild T2D, or a patient with gestational diabetes, if the patient experiences little glucose fluctuation relative to baseline glucose levels (e.g.. a smooth glucose trace around a baseline glucose level), less glucose fluctuation during fasting periods when compared to healthy patients, a meal-dependent post-prandial return to baseline glucose levels after three hours following consumption of a meal, and / or an initial slower glucose level rate of change following consumption of a meal when compared to the glucose rate of change of healthy patients following consumption of a meal, then therapy management engine 114 determines that the root cause of the patient’s determined classification is insulin resistance at block 406.Attorney Docket No.: 0917-PCT01

[0150] In another example, for a patient whose determined classification is a pre-diabetic patient, a patient with mild T2D, or a patient with gestational diabetes, if the patient experiences a high degree of glucose fluctuation around a glucose baseline level, a similar glucose fluctuation during fasting periods when compared to healthy patients, a meal-dependent post-prandial return to baseline glucose levels (e.g., three hours to return to baseline glucose levels following a large meal and two hours to return to baseline glucose levels following a smaller meal), and / or an initial glucose level rate of change following consumption of a meal similar to the glucose level rates of change of healthy patients following consumption of a meal followed by a slower glucose level rate of change when the patient’s endogenous glucose is depleted, therapy management engine 114 determines that the root cause of the patient’s determined classification is insulin deficiency at block 406.

[0151] Following block 404 or optional block 406, therapy management engine 114 proceeds to block 408. At block 408, therapy management engine 114 provides personalized feedback to the patient based on the patient’s diabetic state. In certain embodiments, the feedback to the patient is further based on the root cause optionally determined at block 406. The feedback to the patient includes exercise recommendations, diet recommendations, lifestyle changes, medication recommendations, and continuous glucose monitoring system recommendations. The feedback to the patient based on the patient’s diabetic state of block 408 is described in more detail in reference to blocks 516, 524, 526, 534, 536, 538, 546, and 548.

[0152] In certain embodiments, therapy management engine 114 provides real-time feedback to the patient based on patient input related to the type of food or the type of exercise that resulted in a negative or positive glucose fluctuation and / or glucose metric. For example, therapy management engine 114 can determine if the patient experienced a hypoglycemic event in response to the patient consuming a banana. Therapy management engine 114 may then recommend the patient avoid bananas and / or foods with a similar glycemic profile in the future. In addition to providing feedback related to the immediate cause of various undesirable glucose fluctuations and / or glucose metrics, therapy management engine 114 may request information related to the patient’s overall health (e.g., how much sleep the patient got the previous night, whether the patient is feeling ill, whether the patient started a new medication, etc.) to provide therapy management guidance to improve the patient’s overall health and prevent futureAttorney Docket No.: 0917-PCT01undesirable glucose fluctuations which may contribute to a progression in the patient’s diabetic state.

[0153] Following the personalized feedback at block 408, therapy management engine 114 can return to block 402 to continue monitoring at least the patient’s glucose levels to monitor the progression or regression of the patient’s diabetic state over time. In certain embodiments, therapy management engine 114 can recommend the patient wear a continuous glucose monitoring system (e.g., continuous analyte monitoring system 104) for a specific time within a time period (e.g., one week every two months). The patient may receive glucose monitoring system wear recommendations in the form of a schedule including time periods to wear the continuous glucose monitoring system and time periods to not wear the continuous glucose monitoring system. The recommended time for wearing the glucose monitoring system is based on the patient’s diabetic state.

[0154] If the patient’s diabetic state progresses over subsequent glucose monitoring system wears (e.g., from mild T2D to severe T2D), therapy management engine 114 can recommend the patient wear the glucose monitoring system more frequently within the time period. If the patient’s diabetic state improves over subsequent glucose monitoring system wears (e.g., from severe T2D to mild T2D), therapy management engine 114 can recommend the patient wear the glucose monitoring system less frequently within the time period. In certain embodiments, therapy management engine 114 can instruct the patient to maintain the current recommended wear time if the patient is healthy or improvement is otherwise not expected.

[0155] FIG. 5 illustrates a flow diagram of an example method 500. Method 500 outlines additional examples for classifying a diabetic state of a patient, providing personalized therapy management guidance for managing the patient’s health based on the patient’s diabetic state, and monitoring the progression and / or regression of the patient’s diabetic state over time, as described in reference to FIG.4. Method 500 is described below with reference to FIGs. 1 and 2 and their components.

[0156] Method 500 begins at blocks 502 and 504 at the top of FIG. 5. Generally, blocks 502 and 504 correspond to blocks 402 and 404 of the method 400, respectively.

[0157] At block 502, therapy management engine 114 monitors a patient’s glucose levels over a time period to obtain glucose data. As described in reference to FIG. 4, the time period can beAttorney Docket No.: 0917-PCT01a single wear session or multiple wear sessions during which the patient is wearing a continuous analyte monitoring system (such as continuous analyte monitoring system 104). At block 504, therapy management engine 114 can classify the patient’s diabetic state based on the patient’s glucose data. Therapy management engine 114 classifies the patient’s diabetic state using one of the following classifications at block 504: a healthy patient, a pre-diabetic patient, a patient with mild T2D, a patient with severe T2D, or a patient with gestational diabetes. Each classification will be described below in order.

[0158] If the patient’s baseline glucose level is between 60-100 mg / dL, the patient experiences high glucose fluctuation at baseline, a meal-dependent post-prandial spike below 160 mg / dL that occurs during a time period (following a meal) having a duration that is within a predetermined threshold from a predetermined time value (e.g., 45 minutes), a post-prandial return to baseline that occurs during a time period (following a meal) having a duration that is within a predetermined threshold from a predetermined time value (e.g., 120 minutes) or an identification of a current glucose level that is less than 50% of peak glucose level that occurs during a time period (following a meal) having a duration that is within a predetermined threshold from a predetermined time value (e.g., 120 minutes), and / or a greater negative slope following a meal is observed relative to historical patients with diabetes or pre-diabetes, then the patient can be classified as a healthy patient (e.g., a patient who does not diabetes or pre-diabetes). The method proceeds to block 516.

[0159] At block 516, therapy management engine 114 can provide feedback to the patient on diet recommendations and / or exercise recommendations to maintain the patient’s classification. Therapy management engine 114 can, as part of the diet recommendations, provide therapy management guidance relating to the timing or composition of meals to prevent undesirable glucose fluctuations based on the classification of the patient as a healthy patient, the patient’s known exercise schedule, and / or based on historical data demonstrating positive effects of certain foods on the patient’s glucose metrics. In certain embodiments, therapy management engine 114 can recommend the patient wear a continuous glucose sensor system (e.g., continuous analyte monitoring system 104) for one week every six months to one year to monitor any possible changes in the patient’s classification over time.

[0160] Alternatively, returning to block 504, if the patient’s baseline glucose level is around, or slightly above 100 mg / dL and / or the patient experiences a high meal-dependent post- prandialAttorney Docket No.: 0917-PCT01spike below 200 mg / dL at around 45-60 minutes following a meal, then the patient can be classified as a pre-diabetic patient and the method proceeds to block 518. At block 518, therapy management engine 114 determines whether the patient is insulin resistant or insulin deficient as described in reference to FIG. 4.

[0161] If the patient is classified as an insulin resistant pre-diabetic, therapy management engine 114 proceeds to block 524. At block 524, therapy management engine 114 can recommend the patient complete regular Zone 2 exercise sessions to improve muscle mitochondria and insulin resistance, complete regular resistance training and consume a high protein diet to increase muscle mass, complete an exercise regimen to lose weight, and / or monitor lactate levels to determine improvement or decline in metabolic function. Therapy management engine 114 can further provide therapy management guidance relating to the timing (e.g., before an exercise session) or composition of meals (e.g., avoid meals and foods high carbohydrates) to prevent undesirable glucose fluctuations based on the patient’s classification, the patient’s known exercise schedule, and / or based on historical data demonstrating positive effects of certain foods on the patient’s glucose metrics. In certain embodiments, therapy management engine 114 can recommend the patient wear a continuous glucose sensor system (e.g., continuous analyte monitoring system 104) for one week every three to six months to monitor any possible changes in the patient’s classification over time.

[0162] If the patient was classified as an insulin deficient pre-diabetic, therapy management engine 114 proceeds to block 526. At block 526, therapy management engine 114 can recommend the patient complete a mild exercise session following a meal or an expected increase in glucose to reduce the rise in glucose levels, follow a low carbohydrate diet, and / or consume smaller, more frequent meals. Therapy management engine 114 can further provide therapy management guidance relating to the timing or composition of meals to prevent undesirable glucose fluctuations based on the patient’s classification, the patient’s known exercise schedule, and / or based on historical data demonstrating positive effects of certain foods on the patient’s glucose metrics. Therapy management guidance relating to the timing or composition of meals can include a recommendation to consume meals higher in protein, fiber, or fat to control the magnitude of the patient’s glucose spike following the meal, and / or a recommendation to consume several smaller meals throughout the day. Further, therapy management engine 114 can recommend the patient avoid consuming glucose in liquid form (e.g., orange juice) to further control the magnitude of theAttorney Docket No.: 0917-PCT01patient’s glucose spike following a meal. For both a patient with insulin resistant pre-diabetes and insulin deficient pre-diabetes, therapy management engine 114 can recommend the patient wear a continuous glucose sensor system (e.g., continuous analyte monitoring system 104) for one week every three to six months to monitor any possible changes in the patient’s classification over time.

[0163] Alternatively, returning to block 504, if the patient’s baseline glucose level is between 100-150 mg / dL and / or the patient experiences a meal-dependent post-prandial glucose spike between 150-225 mg / dL at around 60 minutes following a meal, the patient can be classified as a patient with mild T2D and the method proceeds to block 528. At block 528, therapy management engine 114 determines whether the patient is insulin resistant or insulin deficient as described in reference to FIG. 4.

[0164] If the patient is classified as an insulin resistant mild T2D patient, therapy management engine 114 proceeds to block 534. At block 534, therapy management engine 114 can recommend the patient complete regular Zone 2 exercise sessions to improve muscle mitochondrial function and insulin resistance, complete regular resistance training and consume a high protein diet to increase muscle mass, complete an exercise regimen to lose weight, monitor lactate levels to determine improvement or decline in metabolic function, and / or take one or more medications including, for example, glucose lowering medications (e.g., Metformin) and / or GLP-1 agonists. Therapy management engine 114 can further provide therapy management guidance relating to the timing or composition of meals to prevent undesirable glucose fluctuations based on the patient’s classification, the patient’s known exercise schedule, and / or based on historical data demonstrating positive effects of certain foods on the patient’s glucose metrics.

[0165] If the patient was classified as an insulin deficient T2D patient, therapy management engine 114 proceeds to block 536. At block 536, therapy management engine 114 can recommend the patient complete a mild exercise session following a meal or expected increase in glucose to reduce the magnitude of the rise in glucose levels, follow a low carbohydrate diet, consume smaller, more frequent meals, and / or begin an insulin treatment regimen. Therapy management engine 114 can further provide therapy management guidance relating to the timing or composition of meals to prevent undesirable glucose fluctuations based on the patient’s classification, the patient’s known exercise schedule, and / or based on historical data demonstrating positive effects of certain foods on the patient’s glucose metrics.Attorney Docket No.: 0917-PCT01

[0166] For both a patient with insulin resistant mild T2D and insulin deficient mild T2D, therapy management engine 114 can recommend the patient wear a continuous glucose sensor system (e.g., continuous analyte monitoring system 104) for two weeks every two to three months to monitor any possible changes in the patient’s classification over time. If the patient’s diabetes is progressing over subsequent time periods, therapy management engine 114 can instruct the patient to wear a continuous glucose sensor system more frequently.

[0167] Alternatively, returning to block 504, if the patient’s baseline glucose level is above 120 mg / dL, the patient experiences little glucose fluctuation at baseline glucose levels, the patient experiences a meal-dependent post-prandial glucose spike greater than 180 mg / dL at greater than 60 minutes following a meal, the patient experiences a meal-dependent post-prandial return to baseline glucose after four hours or more, and / or the patient experiences a slower return to baseline glucose levels (e.g., smaller negative rate of change of glucose) following a meal than historical patients with mild T2D, then therapy management engine 114 can classify the patient as a patient with severe T2D and the method proceeds to block 538.

[0168] In certain embodiments, the patient can input information related to one or more medications the patient is already taking to manage their diabetes and / or to promote glucose control. If the patient is known to be taking one or more medications known to affect the patient’s glucose levels, one or more of the above metrics described to classify the patient as a severe T2D patient may be altered based on the medication and medication dosage the patient is taking. In the case of severe T2D, insulin resistant and insulin deficient diabetes may present with very similar glucose metrics and symptoms. Therefore, therapy management engine 114 can provide personalized recommendations based on the patient’s diabetic state and glucose metrics, but not based on a determination of whether the patient is insulin resistant or insulin deficient.

[0169] At block 538, therapy management engine 114 can provide therapy management guidance to the patient to complete an exercise and / or diet regimen to lose weight, gain muscle, or improve mitochondrial function, consume a high protein or low carbohydrate diet, monitor lactate levels to determine an improvement or decline in metabolic function, and / or take one or more medications including, for example, glucose lowering medications (e.g., Metformin), GLP-1 agonists, and / or insulin. In certain embodiments, for patients who are obese or severely overweight, therapy management engine 114 can recommend the patient lose weight through aAttorney Docket No.: 0917-PCT01diet regimen prior to beginning an exercise regimen for weight loss. In certain other embodiments, therapy management engine 114 can recommend the obese patient begin on an exercise regimen including exercise types suitable for patients with severe T2D (e.g., swimming, jogging, or biking) in addition to the patient’s diet regimen. Therapy management engine 114 can further provide therapy management guidance relating to the timing or composition of meals to prevent undesirable glucose fluctuations based on the patient’s classification, the patient’s known exercise schedule, and / or based on historical data demonstrating positive effects of certain foods on the patient’s glucose metrics.

[0170] In certain embodiments, therapy management engine 114 can recommend the patient wear a continuous glucose sensor system (e.g., continuous analyte monitoring system 104) for two weeks each month to monitor any possible changes in the patient’s classification over time. If the patient’s diabetic state is progressing over subsequent time periods, therapy management engine 114 can instruct the patient to wear a continuous glucose sensor system more frequently.

[0171] Alternatively, returning to block 504, if the patient is known to be pregnant based on patient profile 118 and / or inputs 130, therapy management engine 114 can determine whether the patient is in their second or third trimester based on patient input, for example. Therapy management engine 114 may continue monitoring glucose data to determine whether a patient has gestational diabetes. If the patient is in their second trimester and experiencing a fasting baseline glucose greater than 95 mg / dL, minor baseline glucose fluctuations not exceeding 140 mg / dL, a meal-dependent post- prandial spike greater than 160 mg / dL during a time period (following a meal) having a duration that is within a predetermined threshold from a predetermined time value (e.g., 60 minutes), a post-prandial return to glucose levels below 120 mg / dL during a time period (following a meal) having a duration that is within a predetermined threshold from a predetermined time value (e.g.. 2 hours), and / or a slightly negative rate of change of glucose levels back to baseline glucose levels following a post-prandial spike, then therapy management engine 114 determines the patient has gestational diabetes and proceeds to block 540.

[0172] If the patient is in their third trimester and experiencing a fasting baseline glucose greater than 92 mg / dL, minor baseline glucose fluctuations not exceeding 120 mg / dL, a mealdependent post-prandial spike greater than 160 mg / dL during a time period (following a meal) having a duration that is within a predetermined threshold from a predetermined time value (e.g.,Attorney Docket No.: 0917-PCT0160 minutes), a post-prandial return to glucose levels below 120 mg / dL during a time period (following a meal) having a duration that is within a predetermined threshold from a predetermined time value (e.g.. 2 hours), and / or a slightly negative rate of change of glucose levels back to baseline glucose levels following a post-prandial spike, then therapy management engine 114 determines the patient has gestational diabetes and proceeds to block 540.

[0173] At block 540, therapy management engine 114 determines whether the patient is insulin resistant or insulin deficient as described in reference to FIG. 4.

[0174] If therapy management engine 114 determines the patient has insulin resistant gestational diabetes at block 540, method 500 proceeds to block 546 to provide recommendations to the patient based on their classification as an insulin resistant gestational diabetic patient.

[0175] At block 546, therapy management engine 114 may recommend the patient begin taking a glucose lowering medication (e.g., Metformin) and / or insulin, consume protein first in a meal, and / or complete a short walk or other low intensity exercise following a meal to reduce the patient’s post-prandial glucose spike. The patient may be instructed to avoid GLP-1 agonists until after pregnancy, as they are contraindicated in pregnancy, and sulfonylureas, which are known to have a negative effect on the size of a baby at birth. Therapy management engine 114 may recommend the patient wear a continuous glucose system (e.g., continuous analyte monitoring system 104) for two weeks each month to monitor any possible changes in the patient’s classification over time, especially during pregnancy. If the patient’s diabetic state is progressing over subsequent time periods, therapy management engine 114 may instruct the patient to wear a continuous glucose sensor system more frequently.

[0176] If therapy management engine 114 determines the patient has insulin deficient gestational diabetes at block 540, method 500 proceeds to block 548 to provide therapy management guidance to the patient based on their classification as an insulin deficient gestational diabetic patient.

[0177] At block 548, therapy management engine 114 may recommend the patient begin taking insulin, consume protein first in a meal, consume smaller, more frequent meals (e.g., more frequent than a typical three times per day meal schedule), and / or complete a short walk or other low intensity exercise following a meal to reduce the patient’s post- prandial glucose spike. The patient may be instructed to avoid GLP-1 agonists, as they are contraindicated in pregnancy, andAttorney Docket No.: 0917-PCT01sulfonyl ureas, which are known to have a negative effect on the size of a baby at birth. Therapy management engine 114 may recommend the patient wear a continuous glucose sensor system (e.g., continuous analyte monitoring system 104) for two weeks each month to monitor any possible changes in the patient’s classification over time, especially during pregnancy. If the patient’s diabetic state is progressing over subsequent time periods, therapy management engine 114 may instruct the patient to wear a continuous glucose sensor system more frequently.

[0178] In each of the above classifications, therapy management engine 114 provides therapy management guidance to the patient, or a patient’s caretaker, family member, and / or the patient’s healthcare provider or insurance company to encourage continuous glucose monitoring based on the recommendation, for example. Additionally, the patient’s actions in response to the recommendations may be verified by the patient via user interface of display device 107. For example, and the patient may be prompted to confirm whether certain actions were taken to comply with the therapy management guidance. In certain embodiments, compliance with the recommendations provided by therapy management engine 114 is verified automatically based on the patient’s monitored glucose levels, additional analyte levels, and / or other non-analyte sensor data (e.g., a heart rate monitor, a wearable blood pressure monitor, an accelerometer sensor on a wearable device such as a watch, fitness tracker, and / or patch, etc.).

[0179] Based on the patient input, and / or analyte and / or non-analyte sensor data, therapy management engine 114 may gauge compliance or non-compliance with the therapy management guidance and this information may be provided to the patient’s caretaker, family member, and / or the patient’s healthcare provider to provide further support and encouragement for the patient to implement the various therapy management guidance to improve the patient’s health.

[0180] In certain embodiments, in any of the above classifications, therapy management engine 114 can determine whether a patient is completing an exercise session or other physical activity (e.g., based on non-analyte data) following consumption of a meal and / or when the patient is experiencing a post-prandial glucose level peak. If therapy management engine 114 determines the patient’s physical activity assisted in lowering the post-prandial glucose peak, therapy management engine 114 may adjust the determination of the magnitude of the post-prandial glucose peak based on the amount the patient’ s physical activity lowered the post-prandial glucose peak.Attorney Docket No.: 0917-PCT01

[0181] As described above relative to each of the above described classifications, therapy management engine 114 may continue to monitor the patient’s glucose levels over multiple periods of continuous glucose sensor system wear to monitor the status of the patient’s classification over time. Therapy management engine 114 recommends a schedule for continuous glucose sensor system wear based on the status of the patient’s classification over time.

[0182] For example, if the patient’s glucose metrics and / or classification are improving at each, or at least two or more, consecutive continuous glucose sensor system wear periods, then therapy management engine 114 recommends the patient wear the continuous glucose sensor system for one week every two months. During subsequent continuous glucose sensor system wear periods, therapy management engine 114 continues to monitor the patient’s glucose metrics and / or classification to provide updated schedule recommendations for continuous glucose sensor system wear. In such an example, if the patient’s glucose metrics and / or classification are improving, therapy management engine 114 updates the patient’s schedule to wear the continuous glucose sensor system for one week every two months.

[0183] Alternatively, if the patient’s glucose metrics and / or classification is progressing indicating a worsening diabetic state with each continuous glucose sensor system use, then therapy management engine 114 recommends the patient wear the continuous glucose sensor system for two weeks out of every two months. During subsequent continuous glucose sensor system wear periods, therapy management engine 114 continues to monitor the patient’s glucose metrics and / or classification to provide updated schedule recommendations for continuous glucose sensor system wear. In such an example, if the patient’s glucose metrics and / or classification are improving, therapy management engine 114 updates the patient’s schedule to wear the continuous glucose sensor system for one week every month.

[0184] FIG.6 is a flow diagram depicting a method 600 for training machine learning models to determine a patient’s diabetic state, provide therapy management guidance based on the patient’s diabetic state, and / or monitor the progression and / or regression of the patient’s diabetic state over time.

[0185] Method 600 begins, at block 602, by training server system, such as training server system 140 illustrated in FIG. 1, retrieving data from historical records database, such as historical records database 112 illustrated in FIG. 1. As mentioned herein, historical records database 112Attorney Docket No.: 0917-PCT01provides a repository of up-to-date information and historical information for patients of a continuous analyte monitoring system and connected mobile health application, such as patients of continuous analyte monitoring system 104 and application 106 illustrated in FIG. 1, as well as data for one or more patients who are not, or were not previously, patients of continuous analyte monitoring system 104 and / or application 106. In certain embodiments, historical records database 112 includes one or more data sets of historical patients who have various diabetic state classifications (e.g., T2D, pre-diabetes, and gestational diabetes) and / or patients who have received various therapy management guidance based on their classification.

[0186] Retrieval of data from historical records database 112 by training server system 140, at block 602, may include the retrieval of all, or any subset of, information maintained by historical records database 112. For example, where historical records database 112 stores information for 100,000 patients (e.g., non-users and users of continuous analyte monitoring system 104 and application 106), data retrieved by training server system 140 to train one or more machine learning models may include information for all 100,000 patients or only a subset of the data for those patients, e.g.. data associated with only 50,000 patients or only data from the last ten years.

[0187] As an illustrative example, at block 602, training server system 140 retrieves information for 100,000 patients with various diabetic state classifications (e.g., healthy patient, pre-diabetic patient, insulin resistant mild T2D patient, insulin deficient mild T2D patient, severe T2D patient, and / or gestational diabetes patient) stored in historical records database 112 to train a model to determine a diabetic state classification of a patient, provide therapy management guidance to the patient based on the patient’s classification, and / or monitor the progression and / or regression of the patient’s diabetic state over time. Each of the 100,000 patients may have a corresponding data record (e.g., based on their corresponding patient profile), stored in historical records database 112. Each patient profile 118 may include information, such as information discussed with respect to FIG. 3.

[0188] The training server system 140 then uses information in each of the records to train an artificial intelligence or ML model (for simplicity referred to as “ML model” herein). Examples of types of information included in a patient’s patient profile were provided above. The information in each of these records may be featurized (e.g., manually or by training server system 140), resulting in features that can be used as input features for training the ML model. ForAttorney Docket No.: 0917-PCT01example, a patient record may include or be used to generate features related to the patient’s demographic information (e.g., an age of a patient, a gender of the patient, etc.), analyte information, such as glucose metrics, non-analyte information, and / or any other data points in the patient record (e.g., inputs 130, metrics 132, etc.). Features used to train the machine learning model(s) may vary in different embodiments.

[0189] In certain embodiments, each historical patient record retrieved from historical records database 112 is further associated with a label indicating a diabetic state classification of the patient (e.g., healthy patient, pre-diabetic patient, insulin resistant mild T2D patient, insulin deficient mild T2D patient, severe T2D patient, and / or gestational diabetes patient). What the record is labeled with would depend on what the model is being trained to predict.

[0190] At block 604, method 600 continues by training server system 140 training one or more machine learning models based on the features and labels associated with the historical patient records. In some embodiments, the training server does so by providing the features as input into a model. This model may be a new model initialized with random weights and parameters, or may be partially or fully pre-trained (e.g., based on prior training rounds). Based on the input features, the model-in-training generates some output. In certain embodiments, the output includes a classification of the patient’s diabetic state, therapy management guidance to manage or improve the patient’s health based on the patient’s classification, a determination of the progression and / or regression of the patient’s diabetic state over time, or similar outputs. Note that the output could be in the form of a classification, therapy management guidance, and / or other types of output.

[0191] In certain embodiments, training server system 140 compares this generated output with the actual label associated with the corresponding historical patient record to compute a loss based on the difference between the actual result and the generated result. This loss is then used to refine one or more internal weights and parameters of the model (e.g., via backpropagation) such that the model learns to determine a diabetic state, provide therapy management guidance to manage or improve the patient’s health based on the patient’s diabetic state, and / or monitor the progression and / or regression of the patient’s diabetic state over time more accurately.

[0192] One of a variety of machine learning algorithms may be used for training the model(s) described above. For example, one of a supervised learning algorithm, a neural network algorithm, a deep neural network algorithm, a deep learning algorithm, etc. may be used.Attorney Docket No.: 0917-PCT01

[0193] At block 606, training server system 140 deploys the trained model(s) to classify a patient’s diabetic state during runtime. In some embodiments, this includes transmitting some indication of the trained model(s) (e.g., a weights vector) that can be used to instantiate the model(s) on another device. For example, training server system 140 can transmit the weights of the trained model(s) to therapy management engine 114, which could execute on display device 107, etc. The model(s) can then be used to classify, in real-time, a patient’s diabetic state using application 106, and / or make other types of recommendations discussed above. In certain embodiments, the training server system 140 continues to train the model(s) in an “online” manner by using input features and labels associated with new patient records.

[0194] Further, similar methods for training illustrated in FIG. 6 using historical patient records may also be used to train models using patient- specific records to create more personalized models for making a determination of the patient’ s diabetic state. For example, a model trained using historical patient records that is deployed for a particular patient, can be further re-trained after deployment. For example, the model can be re-trained after the model is deployed for a specific patient to create a more personalized model for the patient. The more personalized model may be able to more accurately make determinations of the patient’s diabetic state, provide feedback on the progression or improvement of the patient’s diabetic state, and / or provide therapy management guidance based on the patient’s diabetic state and the patient’s own data (as opposed to only historical patient record data), including the patient’s own inputs 130 and metrics 132.

[0195] FIG. 7 is a block diagram depicting a computing device 700 configured to execute a therapy management engine (e.g., therapy management engine 114), according to certain embodiments disclosed herein. Although depicted as a single physical device, in embodiments, computing device 700 can be implemented using virtual device(s), and / or across a number of devices, such as in a cloud environment. As illustrated, computing device 700 includes a processor 705, memory 710, storage 715, a network interface 725, and one or more I / O interfaces 720. In the illustrated embodiment, processor 705 retrieves and executes programming instructions stored in memory 710, as well as stores and retrieves application data residing in storage 715. Processor 705 is generally representative of a single CPU and / or GPU, multiple CPUs and / or GPUs, a single CPU and / or GPU having multiple processing cores, and the like. Memory 710 is generally included to be representative of a random-access memory. Storage 715 may be any combination of disk drives, flash-based storage devices, and the like, and may include fixed and / or removableAttorney Docket No.: 0917-PCT01storage devices, such as fixed disk drives, removable memory cards, caches, optical storage, network attached storage (NAS), or storage area networks (SAN).

[0196] In some embodiments, input and output (I / O) devices 735 (such as keyboards, monitors, etc.) can be connected via the I / O interface(s) 720. Further, via network interface 725, computing device 700 can be communicatively coupled with one or more other devices and components, such as patient database 110. In certain embodiments, computing device 700 is communicatively coupled with other devices via a network, which can include the Internet, local network(s), and the like. The network can include wired connections, wireless connections, or a combination of wired and wireless connections. As illustrated, processor 705, memory 710, storage 715, network interface(s) 725, and I / O interface(s) 720 are communicatively coupled by one or more interconnects 730. In certain embodiments, computing device 700 is representative of display device 107 associated with the patient. In certain embodiments, as discussed above, the display device 107 can include the patient’s laptop, computer, smartphone, and the like. In another embodiment, computing device 700 is a server executing in a cloud environment.

[0197] In the illustrated embodiment, storage 715 includes patient profile 118. Memory 710 includes therapy management engine 114, which itself includes DAM 116.Additional Considerations

[0198] The methods disclosed herein comprise one or more steps or actions for achieving the methods. The method steps and / or actions may be interchanged with one another without departing from the scope of the claims. In other words, unless a specific order of steps or actions is specified, the order and / or use of specific steps and / or actions may be modified without departing from the scope of the claims.

[0199] As used herein, a phrase referring to “at least one of’ a list of items refers to any combination of those items, including single members. As an example, “at least one of: a, b, or c” is intended to cover a, b, c, a-b, a-c, b-c, and a-b-c, as well as any combination with multiples of the same element (e.g., a-a, a-a-a, a-a-b, a-a-c. a-b-b, a-c-c. b-b, b-b-b, b-b-c, c-c. and c-c-c or any other ordering of a, b, and c).Attorney Docket No.: 0917-PCT01

[0200] The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects. Thus, the claims are not intended to be limited to the aspects shown herein, but is to be accorded the full scope consistent with the language of the claims, wherein reference to an element in the singular is not intended to mean “one and only one” unless specifically so stated, but rather “one or more.” Unless specifically stated otherwise, the term “some” refers to one or more. All structural and functional equivalents to the elements of the various aspects described throughout this disclosure that are known or later come to be known to those of ordinary skill in the art are expressly incorporated herein by reference and are intended to be encompassed by the claims. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether such disclosure is explicitly recited in the claims. No claim element is to be construed under the provisions of 35 U.S.C. § 112(f) unless the element is expressly recited using the phrase “means for” or, in the case of a method claim, the element is recited using the phrase “step for.”

[0201] While various examples of the invention have been described above, it should be understood that they have been presented by way of example only, and not by way of limitation. Likewise, the various diagrams may depict an example architectural or other configuration for the disclosure, which is done to aid in understanding the features and functionality that can be included in the disclosure. The disclosure is not restricted to the illustrated example architectures or configurations, but can be implemented using a variety of alternative architectures and configurations. Additionally, although the disclosure is described above in terms of various example examples and aspects, it should be understood that the various features and functionality described in one or more of the individual examples are not limited in their applicability to the particular example with which they are described. They instead can be applied, alone or in some combination, to one or more of the other examples of the disclosure, whether or not such examples are described, and whether or not such features are presented as being a part of a described example. Thus the breadth and scope of the present disclosure should not be limited by any of the above-described example examples.

[0202] All references cited herein are incorporated herein by reference in their entirety. To the extent publications and patents or patent applications incorporated by reference contradict theAttorney Docket No.: 0917-PCT01disclosure contained in the specification, the specification is intended to supersede and / or take precedence over any such contradictory material.

[0203] Unless otherwise defined, all terms (including technical and scientific terms) are to be given their ordinary and customary meaning to a person of ordinary skill in the art, and are not to be limited to a special or customized meaning unless expressly so defined herein.

[0204] Terms and phrases used in this application, and variations thereof, especially in the appended claims, unless otherwise expressly stated, should be construed as open ended as opposed to limiting. As examples of the foregoing, the term ‘including’ should be read to mean ‘including, without limitation,’ ‘including but not limited to,’ or the like; the term ‘comprising’ as used herein is synonymous with ‘including,’ ‘containing,’ or ‘characterized by.’ and is inclusive or open-ended and does not exclude additional, unrecited elements or method steps; the term ‘having’ should be interpreted as ‘having at least;’ the term ‘includes’ should be interpreted as ‘includes but is not limited to;’ the term ‘example’ is used to provide example instances of the item in discussion, not an exhaustive or limiting list thereof; adjectives such as ‘known’, ‘normal’, ‘standard’, and terms of similar meaning should not be construed as limiting the item described to a given time period or to an item available as of a given time, but instead should be read to encompass known, normal, or standard technologies that may be available or known now or at any time in the future; and use of terms like ‘preferably.’ ‘preferred,’ ‘desired,’ or ‘desirable.’ and words of similar meaning should not be understood as implying that certain features are critical, essential, or even important to the structure or function of the invention, but instead as merely intended to highlight alternative or additional features that may or may not be utilized in a particular example of the invention. Likewise, a group of items linked with the conjunction ‘and’ should not be read as requiring that each and every one of those items be present in the grouping, but rather should be read as ‘and / or’ unless expressly stated otherwise. Similarly, a group of items linked with the conjunction ‘or’ should not be read as requiring mutual exclusivity among that group, but rather should be read as ‘and / or’ unless expressly stated otherwise.

[0205] The term “comprising as used herein is synonymous with “including,” “containing,” or “characterized by” and is inclusive or open-ended and does not exclude additional, unrecited elements or method steps.Attorney Docket No.: 0917-PCT01

[0206] All numbers expressing quantities of ingredients, reaction conditions, and so forth used in the specification are to be understood as being modified in all instances by the term ‘about.’ Accordingly, unless indicated to the contrary, the numerical parameters set forth herein are approximations that may vary depending upon the desired properties sought to be obtained. At the very least, and not as an attempt to limit the application of the doctrine of equivalents to the scope of any claims in any application claiming priority to the present application, each numerical parameter should be construed in light of the number of significant digits and ordinary rounding approaches.

[0207] Furthermore, although the foregoing has been described in some detail by way of illustrations and examples for purposes of clarity and understanding, it is apparent to those skilled in the art that certain changes and modifications may be practiced. Therefore, the description and examples should not be construed as limiting the scope of the invention to the specific examples and examples described herein, but rather to also cover all modification and alternatives coming with the true scope and spirit of the invention.Example Embodiments

[0208] Clause 1 : A method for generating personalized therapy management guidance for a user, comprising: monitoring continuous analyte data of a user during a first time period to obtain continuous analyte metrics of the user during the first time period, the continuous analyte data comprising at least glucose data and the continuous analyte metrics comprising at least glucose metrics; determining a diabetic state classification of the user based on the analyte metrics, the diabetic state classification indicating a diabetic state of the user; and generating personalized therapy management guidance for the user based on the determined diabetic state classification.

[0209] Clause 2: The method of Clause 1, wherein monitoring the continuous analyte data of the user during the first time period to obtain the continuous analyte metrics during the first time period comprises: receiving the continuous analyte data and processing the continuous analyte data to identify analyte metrics of the user.

[0210] Clause 3: The method of any combination of Clauses 1-2, wherein: the continuous analyte data further comprises at least one of lactate data or ketone data; and the analyte metrics further comprise at least one of lactate metrics or ketone metrics.Attorney Docket No.: 0917-PCT01

[0211] Clause 4: The method of any combination of Clauses 1-3, wherein determining the diabetic state classification comprises applying a rules-based model that evaluates one or more glucose metrics selected from: a baseline glucose level; a post-prandial glucose excursion; a rate of return to baseline following a glucose excursion; a glucose response to light exercise; or glucose variability during fasting.

[0212] Clause 5: The method of Clause 4, wherein determining the diabetic state classification further comprises applying the rules-based model that evaluates one or more non-analyte user inputs selected from: an age of the user; or a disease diagnosis of the user.

[0213] Clause 6: The method of any combination of Clauses 1-5, wherein determining the diabetic state classification comprises applying an artificial intelligence or machine learning model trained on labeled datasets of users with known diabetic state classifications.

[0214] Clause 7: The method of any combination of Clauses 1-6, further comprising: identifying one or more periods of glucose data impacted by external factors, the external factors comprising at least one of alcohol consumption, short-term illness, acute flare-ups of chronic disease, or administration of medication; and excluding or down-weighting impacted glucose data when determining the diabetic state classification.

[0215] Clause 8: The method of Clause 7, wherein identifying the one or more periods of glucose data impacted by the external factors comprises: receiving user input describing at least one of recent alcohol consumption, recent illness, or recent medication administration; or detecting patterns in the continuous analyte data indicating the external factor.

[0216] Clause 9: The method of any combination of Clauses 7-8. further comprising pausing the determination of the diabetic state classification until a trigger event occurs, the trigger event comprising at least one of: a lapse of a threshold time period; a determination that glucose levels of the user returned to baseline; or a determination that the external factor has resolved.

[0217] Clause 10: The method of any combination of Clauses 1-9, further comprising: updating the determined diabetic state classification upon lapse of a second time period, wherein updating the determined diabetic state classification comprises: receiving additional continuous analyte data associated with the user during the second time period; processing the additional continuous analyte data to identify new analyte metrics for the user, the new analyte metricsAttorney Docket No.: 0917-PCT01comprising at least new glucose metrics; and adjusting the diabetic state classification when a change in the diabetic state is detected, the change in the diabetic state determined based, at least in part, on the new glucose metrics.

[0218] Clause 11: The method of Clause 10, wherein detecting the change in the diabetic state comprises at least one of: identifying progression from a less severe diabetic state to a more severe diabetic state; or identifying regression from a more severe to a less severe diabetic state; or identifying a change to a compounded classification of insulin resistance and insulin dependence.

[0219] Clause 12: The method of any combination of Clauses 1-11, further comprising: determining a root cause for the diabetic state of the user.

[0220] Clause 13: The method of Clause 12, wherein determining the root cause for the diabetic state is performed when the diabetic state classification comprises pre-diabetes, mild type 2 diabetes, or gestational diabetes.

[0221] Clause 14: The method of any combination of Clauses 12-13, wherein determining the root cause comprises: identifying an insulin resistance of the user when the glucose data demonstrates low fasting variability, delayed return to baseline post-prandially, and slower initial glucose rate of change after meals relative to healthy subjects.

[0222] Clause 15: The method of any combination of Clauses 12-14, wherein determining the root cause comprises: identifying an insulin deficiency of the user when the glucose data demonstrates high fasting variability, faster early meal response with slower subsequent decline, and return to baseline within about two to three hours after meals.

[0223] Clause 16: The method of any combination of Clauses 1-15, wherein the personalized therapy management guidance comprises a recommendation for at least one of: an exercise program; a dietary change; a lifestyle modification; a medication adjustment; or a continuous glucose monitoring wear schedule.

[0224] Clause 17: The method of Clause 16, wherein the continuous glucose monitoring wear schedule comprises a suggestion for more frequent sensor wear upon detection of a progression of the diabetic state, and less frequent wear upon detection of a regression of the diabetic state.

[0225] Clause 18: The method of any combination of Clauses 1-17, further comprising: receiving user input regarding at least one of a consumption of a meal or performance of exercise,Attorney Docket No.: 0917-PCT01wherein the personalized therapy management guidance is based, at least in part, on the user input and associated glucose responses to the consumption of the meal or the performance of the exercise.

[0226] Clause 19: The method of any combination of Clauses 1-18, further comprising: storing historical diabetic state classifications and analyte metrics of the user; and generating updated personalized therapy management guidance comprising at least an indication of a progression or a regression of the diabetic state of the user over time.

[0227] Clause 20: The method of any combination of Clauses 1-19, wherein the diabetic state classification comprises one of: a healthy user; a pre-diabetic user; a mild type 2 diabetes (T2D) user; a severe T2D user; or a gestational diabetes user.

[0228] Clause 21: The method of Clause 20, wherein the healthy user classification is determined based, at least in part, on the identification of at least one of the following glucose metrics: a baseline glucose level between 60-100 mg / dL; high glucose fluctuation at baseline glucose levels; a meal-dependent post-prandial spike below 160 mg / dL that occurs within about 45 minutes following consumption of a meal; a post-prandial return to baseline that occurs within about 120 minutes following consumption of the meal; a glucose level that is less than 50% of peak glucose level that occurs within about 120 minutes following consumption of the meal; or a greater negative slope of glucose levels following consumption of the meal relative to historical patients with diabetes or pre-diabetes.

[0229] Clause 22: The method of any combination of Clauses 20-21, wherein: the diabetic state classification comprises the healthy user classification; and the personalized therapy management guidance comprises at least one of a diet recommendation or an exercise recommendation to maintain the healthy user classification.

[0230] Clause 23: The method of any combination of Clause 22, wherein the diet recommendation comprises a recommendation relating to the timing or composition of meals to prevent undesirable glucose fluctuations.

[0231] Clause 24: The method of any combination of Clauses 20-23, wherein: the diabetic state classification comprises the healthy user classification; and the personalized therapy management guidance comprises a recommendation to wear a continuous glucose sensor systemAttorney Docket No.: 0917-PCT01for one week every six months to one year to monitor changes in the user’s classification over time.

[0232] Clause 25: The method of Clause 20, wherein the pre-diabetic user classification is determined based, at least in part, on the identification of at least one of the following glucose metrics: a baseline glucose level of about 100 mg / dL or more; or a meal-dependent post-prandial spike below 200 mg / dL that occurs between about 45 minutes and 60 minutes following consumption of a meal.

[0233] Clause 26: The method of Clause 25, wherein: the diabetic state classification comprises the pre-diabetic user classification; and the method further comprises: determining whether the user is insulin-resistant or insulin-deficient based, at least in part on, the analyte metrics.

[0234] Clause 27: The method of Clause 26, wherein: the user is determined to be insulinresistant; and the personalized therapy management guidance comprises at least one of: a recommendation to perform Zone 2 exercise to improve muscle mitochondria and insulin resistance; a recommendation to perform resistance training and consume a high protein diet to increase muscle mass; a recommendation to complete an exercise regimen to lose weight; or a recommendation to monitor lactate levels to determine improvement or decline in metabolic function.

[0235] Clause 28: The method of Clause 26, wherein: the user is determined to be insulindeficient; and the personalized therapy management guidance comprises at least one of: a recommendation to perform a mild exercise session following a meal or an expected increase in glucose to retard a rise in glucose levels; a recommendation to follow a low carbohydrate diet; a recommendation to follow a high protein, high fiber, or high fat diet; or a recommendation to consume smaller, more frequent meals.

[0236] Clause 29: The method of any combination of Clauses 26-29, wherein personalized therapy management guidance comprises a diet recommendation relating to the timing or composition of meals to prevent undesirable glucose fluctuations.

[0237] Clause 30: The method of any combination of Clauses 26-29, wherein the personalized therapy management guidance comprises a recommendation to wear a continuous glucose sensorAttorney Docket No.: 0917-PCT01system for one week every three months to six months to monitor changes in the user’s classification over time.

[0238] Clause 31: The method of Clause 20, wherein the mild T2D user classification is determined based, at least in part, on the identification of at least one of the following glucose metrics: a baseline glucose level between 100-150 mg / dL; or a meal-dependent post-prandial spike between 150-225 mg / dL that occurs about 60 minutes following consumption of a meal.

[0239] Clause 32: The method of Clause 31, wherein: the diabetic state classification comprises the mild T2D user classification; and the method further comprises: determining whether the user is insulin-resistant or insulin-deficient based, at least in part on, the analyte metrics.

[0240] Clause 33: The method of Clause 32, wherein: the user is determined to be insulinresistant; and the personalized therapy management guidance comprises at least one of: a recommendation to perform Zone 2 exercise to improve muscle mitochondria and insulin resistance; a recommendation to perform resistance training and consume a high protein diet to increase muscle mass; a recommendation to complete an exercise regimen to lose weight; a recommendation to monitor lactate levels to determine improvement or decline in metabolic function; or a recommendation to administer one or more medications.

[0241] Clause 34: The method of Clause 32, wherein: the user is determined to be insulindeficient; and the personalized therapy management guidance comprises at least one of: a recommendation to perform a mild exercise session following a meal or an expected increase in glucose to retard a rise in glucose levels; a recommendation to follow a low carbohydrate diet; a recommendation to consume smaller, more frequent meals; or a recommendation to begin an insulin treatment regimen.

[0242] Clause 35: The method of any combination of Clauses 31-34, wherein personalized therapy management guidance comprises a diet recommendation relating to the timing or composition of meals to prevent undesirable glucose fluctuations.

[0243] Clause 36: The method of any combination of Clauses 31-34, wherein the personalized therapy management guidance comprises a recommendation to wear a continuous glucose sensorAttorney Docket No.: 0917-PCT01system for two weeks every two months to three months to monitor changes in the user’s classification over time.

[0244] Clause 37: The method of Clause 20, wherein the severe T2D user classification is determined based, at least in part, on the identification of at least one of the following glucose metrics: a baseline glucose level above 120 mg / dL; few glucose fluctuation at baseline glucose levels; a meal-dependent post-prandial spike greater than 180 mg / dL that occurs more than about 60 minutes following consumption of a meal; a meal-dependent post-prandial return to baseline glucose levels after four hours or more; or smaller negative rate of change of glucose following a meal relative to historical patients with mild T2D.

[0245] Clause 38: The method of Clause 37, wherein: the diabetic state classification comprises the severe T2D user classification; and the personalized therapy management guidance comprises at least one of: a recommendation to perform an exercise regimen to lose weight, gain muscle, or improve mitochondrial function; a recommendation to begin a diet regimen to lose weight, gain muscle, or improve mitochondrial function; a recommendation to follow a low carbohydrate and high protein diet; a recommendation to monitor lactate levels to monitor improvement or regression in metabolic function; or a recommendation to begin a medication regimen.

[0246] Clause 39: The method of any combination of Clauses 37-38, wherein personalized therapy management guidance comprises a diet recommendation relating to the timing or composition of meals to prevent undesirable glucose fluctuations.

[0247] Clause 40: The method of any combination of Clauses 37-39, wherein the personalized therapy management guidance comprises a recommendation to wear a continuous glucose sensor system for two weeks every month to monitor changes in the user’s classification over time.

[0248] Clause 41: The method of Clause 20, further comprising: receiving one or more nonanalyte user inputs indicating that the user is pregnant and in a second trimester or a third trimester of pregnancy.

[0249] Clause 42: The method of Clause 41, wherein the gestational diabetes user classification is determined based, at least in part, on: the one or more non-analyte user inputs indicating that the user is pregnant and in the second trimester; and the identification of at leastAttorney Docket No.: 0917-PCT01one of the following glucose metrics: a fasting baseline glucose level greater than 95 mg / dL; minor baseline glucose fluctuations not exceeding 140 mg / dL; a meal-dependent post-prandial spike greater than 160 mg / dL that occurs about 60 minutes following consumption of a meal; a postprandial return to glucose levels below 120 mg / dL that occurs within about 120 minutes following consumption of a meal; or a slightly negative rate of change of glucose levels back to baseline glucose levels following a post-prandial spike.

[0250] Clause 43: The method of any combination of Clauses 41-42, wherein: the diabetic state classification comprises the gestational diabetes user classification; and the method further comprises: determining whether the user is insulin-resistant or insulin-deficient based, at least in part on, the analyte metrics.

[0251] Clause 44: The method of Clause 43, wherein: the user is determined to be insulinresistant; and the personalized therapy management guidance comprises at least one of: a recommendation to consume protein first in a meal; a recommendation to perform low intensity exercise following a meal to mitigate a post-prandial glucose spike; or a recommendation to begin one or more medications and / or insulin for lowering glucose levels.

[0252] Clause 45: The method of Clause 43. wherein: the user is determined to be insulindeficient; and the personalized therapy management guidance comprises at least one of: a recommendation to begin an insulin regimen; a recommendation to consume smaller, more frequent meals; or a recommendation to perform low intensity exercise following a meal to mitigate a post-prandial glucose spike.

[0253] Clause 46 The method of any combination of Clauses 42-45, wherein personalized therapy management guidance comprises a diet recommendation relating to the timing or composition of meals to prevent undesirable glucose fluctuations.

[0254] Clause 47: The method of any combination of Clauses 42-46, wherein the personalized therapy management guidance comprises a recommendation to wear a continuous glucose sensor system for two weeks every month to monitor changes in the user’s classification over time.

[0255] Clause 48: The method of any combination of Clauses 1-47, further comprising: monitoring new glucose data during subsequent monitoring periods; and adjusting the classification when a change in the diabetic state is detected.Attorney Docket No.: 0917-PCT01

[0256] Clause 49: The method of any combination of Clauses 1-48, further comprising: receiving additional continuous analyte data associated with the user during a second time period; processing the additional continuous analyte data to identify new analyte metrics for the user, the new analyte metrics comprising at least new glucose metrics; determining whether the diabetic state of the user is progressing or regressing based, at least in part, on the new analyte metrics; and generating an updated recommendation for a continuous glucose monitoring wear schedule based, at least in part, on the diabetic state of the user progressing or regressing.

[0257] Clause 50: An apparatus, comprising: at least one memory comprising executable instructions; and at least one processor configured to execute the executable instructions and cause the apparatus to perform a method in accordance with any combination of Clauses 1-49.

[0258] Clause 51: An apparatus, comprising means for performing a method in accordance with any combination of Clauses 1-49.

[0259] Clause 52: A non-transitory computer- readable medium comprising executable instructions that, when executed by at least one processor of an apparatus, cause the apparatus to perform a method in accordance with any combination of Clauses 1-49.

[0260] Clause 53: A computer program product embodied on a computer- readable storage medium comprising code for performing a method in accordance with any combination of Clauses 1-49.

Claims

Attorney Docket No.: 0917-PCT01What is claimed is:

1. A therapy management system, comprising:one or more memories comprising executable instructions; andone or more processors in data communication with the one or more memories and configured to execute the executable instructions to:monitor at least glucose concentration levels of a patient during a first time period using one or more glucose sensors to obtain glucose data;determine a diabetic state classification of the patient based on the glucose data, the diabetic state classification indicating a diabetic state of the patient, wherein the diabetic state classification comprises one of:a healthy classification;a pre-diabetic classification;a mild type 2 diabetes (T2D) classification;a severe T2D classification; ora gestational diabetes classification; andprovide personalized feedback to the patient based on the determined diabetic state classification, wherein the personalized feedback comprises a recommendation for at least one of:an exercise session;a dietary change;a lifestyle modification;a medication adjustment; ora continuous glucose monitoring wear schedule.

2. The therapy management system of claim 1, wherein the glucose data comprises at least one of the following glucose metrics:a baseline glucose level;a post-prandial glucose excursion;a rate of return to baseline following a glucose excursion;a glucose response to light exercise; orAttorney Docket No.: 0917-PCT01glucose variability during fasting.

3. The therapy management system of claim 1 , wherein the one or more processors are further configured to execute the executable instructions to:Monitor at least glucose concentration levels of the patient during a second time period to obtain new glucose data; andadjust the diabetic state classification when a change in the diabetic state is detected, the change in the diabetic state determined based, at least in part, on the new glucose data.

4. The therapy management system of claim 3, wherein detecting the change in the diabetic state comprises at least one of:identifying progression from a less severe diabetic state to a more severe diabetic state; or identifying regression from a more severe to a less severe diabetic state; or identifying a change to a compounded classification of insulin resistance and insulin dependence.

5. The therapy management system of claim 1, wherein:the diabetic state classification comprises one of the pre-diabetic classification, the mild type 2 diabetes (T2D) classification, the severe T2D classification, or the gestational diabetes classification; andthe one or more processors are further configured to execute the executable instructions to determine a root cause for the diabetic state of the patient.

6. The therapy management system of claim 1, wherein the healthy classification is determined based, at least in part, on the identification of at least one of the following glucose metrics in the glucose data:a baseline glucose level between 60-100 mg / dL;high glucose fluctuation at baseline glucose levels;a meal-dependent post-prandial spike below 160 mg / dL that occurs within about 45 minutes following consumption of a meal;Attorney Docket No.: 0917-PCT01a post-prandial return to baseline that occurs within about 120 minutes following consumption of the meal;a glucose level that is less than 50% of peak glucose level that occurs within about 120 minutes following consumption of the meal; ora greater negative slope of glucose levels following consumption of the meal relative to historical patients with diabetes or pre-diabetes.

7. The therapy management system of claim 1, wherein the pre-diabetes classification is determined based, at least in part, on the identification of at least one of the following glucose metrics in the glucose data:a baseline glucose level of about 100 mg / dL or more: ora meal-dependent post-prandial spike below 200 mg / dL that occurs between about 45 minutes and 60 minutes following consumption of a meal.

8. The therapy management system of claim 1, wherein the mild T2D classification is determined based, at least in part, on the identification of at least one of the following glucose metrics in the glucose data:a baseline glucose level between 100-150 mg / dL; ora meal-dependent post-prandial spike between 150-225 mg / dL that occurs about 60 minutes following consumption of a meal.

9. The therapy management system of claim 1, wherein the severe T2D classification is determined based, at least in part, on the identification of at least one of the following glucose metrics in the glucose data:a baseline glucose level above 120 mg / dL;few glucose fluctuation at baseline glucose levels;a meal-dependent post-prandial spike greater than 180 mg / dL that occurs more than about 60 minutes following consumption of a meal;a meal-dependent post-prandial return to baseline glucose levels after four hours or more; orAttorney Docket No.: 0917-PCT01smaller negative rate of change of glucose following a meal relative to historical patients with mild T2D.

10. The therapy management system of claim 1, wherein the gestational diabetes classification is determined based, at least in part, on:the one or more non-analyte patient inputs indicating that the patient is pregnant and in the second trimester; andthe identification of at least one of the following glucose metrics in the glucose data: a fasting baseline glucose level greater than 95 mg / dL;minor baseline glucose fluctuations not exceeding 140 mg / dL;a meal-dependent post-prandial spike greater than 160 mg / dL that occurs about 60 minutes following consumption of a meal;a post-prandial return to glucose levels below 120 mg / dL that occurs within about 120 minutes following consumption of a meal; ora slightly negative rate of change of glucose levels back to baseline glucose levels following a post-prandial spike.

11. The therapy management system of claim 1, wherein:the diabetic state classification comprises the pre-diabetes classification, the mild T2D diabetes classification, or the gestational diabetes classification; andwherein the one or more processors are further configured to execute the executable instructions to:determine whether the patient is insulin-resistant or insulin-deficient based, at least in part on, the glucose metrics.

12. The therapy management system of claim 1 , wherein the one or more processors are further configured to execute the executable instructions to:identify a period of glucose data impacted by external factors, the external factors comprising at least one of alcohol consumption, short-term illness, acute flare-ups of chronic disease, or administration of medication: andAttorney Docket No.: 0917-PCT01excluding or down-weighting impacted glucose data when determining the diabetic state classification.

13. The therapy management system of claim 12, wherein identifying the period of glucose data impacted by the external factors comprises:receiving patient input describing at least one of recent alcohol consumption, recent illness, or recent medication administration; ordetecting patterns in the continuous analyte data indicating the external factor.

14. The therapy management system of claim 12, wherein the one or more processors are further configured to execute the executable instructions to pause the determination of the diabetic state classification until a trigger event occurs, the trigger event comprising at least one of:a lapse of a threshold time period;a determination that glucose levels of the patient returned to baseline; ora determination that the external factor has resolved.

15. A method for generating personalized feedback for a patient, comprising:monitoring at least glucose concentration levels of a patient during a first time period using one or more glucose sensors to obtain glucose data;determining a diabetic state classification of the patient based on the glucose data, the diabetic state classification indicating a diabetic state of the patient, wherein the diabetic state classification comprises one of:a healthy classification;a pre-diabetic classification;a mild type 2 diabetes (T2D) classification;a severe T2D classification; ora gestational diabetes classification; andproviding personalized feedback to the patient based on the determined diabetic state classification, wherein the personalized feedback comprises a recommendation for at least one of:an exercise session;a dietary change;Attorney Docket No.: 0917-PCT01a lifestyle modification;a medication adjustment; ora continuous glucose monitoring wear schedule.

Images