Modeling target for continuous glucose monitoring metrics for glycemic control

Abstract

Disclosed herein are techniques for blood glucose management. In one example, a processor-implemented method includes receiving an input of a target value for a first continuous glucose monitoring (CGM) metric; estimating a target value of at least a second CGM metric corresponding to the target value of the first CGM metric; and providing the estimated target value of at least the second CGM metric to the user. In some examples, the processor-implemented method further includes determining that the estimated target value of at least the second CGM metric satisfies a predetermined criterion; and configuring an insulin delivery system based on the target value of the first CGM metric.

Description

[0001] Cross-reference to related applications

[0002] This application claims the benefit and priority of U.S. Provisional Application No. 63 / 648,352, filed May 16, 2024, entitled “Modeling targets of continuous glucose monitoring measures for glucose control,” which has been assigned to the assignee of this application and is hereby incorporated in its entirety by reference for all purposes. Technical Field

[0003] This disclosure generally relates to blood glucose management. Background Technology

[0004] In a healthy person, the pancreas will produce insulin and release it into the bloodstream in response to elevated blood sugar levels. More specifically, beta cells (β-cells) in the pancreas produce insulin as needed and secrete it into the bloodstream. If beta cells lose their ability to function or die (a condition known as type 1 diabetes), an additional infusion delivery device, such as a syringe, pen, pump, or blood glucose management system, may be needed to deliver insulin to the body of a person with diabetes to maintain health or life.

[0005] Some blood glucose management systems can be closed-loop systems, which may include automatic or semi-automatic controls to deliver insulin to the patient via a pump. Some blood glucose management systems may require manual insulin administration based on a dose determined by an insulin calculator. The timing and amount of insulin delivery to the patient can be controlled, based on, for example, the amount of carbohydrates the patient ingests and / or real-time measurements of glucose levels by a glucose sensor such as a continuous glucose monitor (CGM). In addition to delivering insulin, some blood glucose management systems may also deliver glucose and / or glucagon to control the patient's blood glucose levels (e.g., in the context of hypoglycemia). Summary of the Invention

[0006] This disclosure generally relates to blood glucose management. More specifically, the techniques disclosed herein relate to systems and methods for determining appropriate targets for continuous glucose monitoring (CGM) metrics to guide automated insulin delivery (AID) systems for blood glucose control. The techniques disclosed herein can be practiced in various ways, such as methods implemented using servers, user devices, processors, systems including one or more processors and one or more processor-readable media, and / or one or more (non-transitory) processor-readable media.

[0007] According to certain embodiments, a method can include receiving an input of a target value for a first continuous glucose monitoring (CGM) metric; estimating a target value for at least a second CGM metric corresponding to the target value for the first CGM metric; and providing the estimated target value for at least the second CGM metric to a user.

[0008] According to certain embodiments, a method can include using a classifier to classify continuous glucose monitoring (CGM) data samples that satisfy and do not satisfy a target value for a first CGM metric as samples that satisfy or do not satisfy each threshold target of a plurality of threshold targets for a second CGM metric; determining a true positive rate (TPR) and a false positive rate (FPR) of the classifier for each threshold target of the plurality of threshold targets for the second CGM metric; and identifying a target value for at least the second CGM metric based on the TPR and the FPR of the classifier for each threshold target of the plurality of threshold targets for the second CGM metric.

[0009] This summary is neither intended nor should it be construed to be used to identify key or essential features of the claimed subject matter. The subject matter is better understood from a consideration of the entire description and the proper constructions of the scope of the claims. The foregoing and other features and examples are described in more detail below in the specification, claims and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0010] The foregoing and other aspects and features of the present disclosure will become more apparent from the following detailed description, taken in conjunction with the accompanying drawings, in which like reference characters refer to like elements throughout.

[0011] Figure 1 An example of a blood glucose level management system is illustrated in accordance with certain embodiments.

[0012] Figure 2 is a block diagram of an example of a blood glucose level management system in accordance with certain embodiments.

[0013] Figure 3 An example of a continuous glucose monitoring (CGM) system is illustrated in accordance with certain embodiments.

[0014] Figure 4A and Figure 4B An example of an integrated glucose sensor device is depicted in accordance with certain embodiments.

[0015] Figure 5 An example of an insulin delivery device is depicted in accordance with some embodiments.

[0016] Figures 6A to 6C An example of a continuous glucose monitoring result is illustrated.

[0017] Figure 7 Examples of recommended targets for in-range time metrics for evaluating a blood glucose control system are illustrated.

[0018] Figure 8 A summary of the above examples of CGM datasets for determining targets for CGM metrics based on targets for another CGM metric is shown in accordance with certain embodiments.

[0019] Figure 9 Another summary of examples of CGM datasets for determining targets for CGM metrics based on targets for another CGM metric is shown in accordance with certain embodiments.

[0020] Figure 10 A diagram including examples illustrating receiver operating characteristic (ROC) curves.

[0021] Figure 11 A table is included illustrating examples of determined target values for certain in-range time metrics that can correspond to different GMI target values in accordance with certain embodiments.

[0022] Figure 12A and Figure 12B Results of Pearson correlation analysis of six in-range time metrics and GMI metrics for users 15 years old or younger and users over 15 years old, respectively, are shown.

[0023] Figure 12C and Figure 12D Results of principal component analysis (PCA) analysis of six in-range time metrics and GMI metrics for users 15 years old or younger and users over 15 years old, respectively, are shown.

[0024] Figure 13 A table is included showing examples of estimated target results for TB54 associated with various TB70 targets in accordance with certain embodiments.

[0025] Figure 14 A flowchart including examples illustrating methods for determining target values for CGM metrics that correspond to target values for another CGM target in accordance with certain embodiments is included.

[0026] Figure 15 A flowchart including examples illustrating processor-implemented methods for providing target values for CGM metrics that correspond to target values for another CGM target to a user in accordance with certain embodiments is included.

[0027] Figure 16 is a block diagram of an example of a computer system that can be used to implement some embodiments described herein.

[0028] The accompanying drawings illustrate embodiments of the present disclosure for illustrative purposes only. Those skilled in the art will readily recognize from the following description that alternative embodiments of the illustrated structures and methods may be employed without departing from the principles of the present disclosure or the benefits claimed.

[0029] In the accompanying drawings, similar parts and / or features may have the same reference numerals. Additionally, various parts of the same type can be distinguished by adding a dash after the reference numerals and a second reference numeral to differentiate similar parts. If only the first reference numeral is used in the specification, the description applies to any of the similar parts having the same first reference numeral, regardless of the second reference numeral. Detailed Implementation

[0030] The techniques disclosed herein generally relate to blood glucose management. More specifically, the techniques disclosed herein relate to systems and methods for determining appropriate targets for continuous glucose monitoring (CGM) metrics to guide automated insulin delivery (AID) systems for blood glucose control.

[0031] Diabetes is a disease of the patient's glucose regulation system, in which the body's naturally produced insulin may be insufficient to control glucose levels in the bloodstream due to inadequate insulin production and / or insulin resistance. Therefore, people with diabetes may need to receive insulin from a pump or another delivery device (such as an injection or infusion device, e.g., a syringe) to control glucose levels. A typical treatment routine for people with diabetes generally includes a basal insulin dose and a bolus insulin dose. Basal insulin (also known as background insulin) consists of a continuous or constant release of small amounts of insulin to maintain a consistent blood glucose level over a long period. A bolus insulin dose may be taken specifically before, during, or after meals, or at other times when blood glucose levels may rise rapidly.

[0032] The insulin dose to be delivered can be determined based on, for example, the carbohydrate count of the meal and / or the patient's glucose level as measured using a blood glucose monitor (such as a finger-prick glucose meter or a continuous glucose monitoring (CGM) sensor). In one example, to counteract an increase in the patient's blood glucose level due to consumption of food (or drink), a dose of insulin (called a dietary bolus) can be delivered to the patient before, during, or shortly after a meal. The insulin dose can be determined using an insulin calculator (e.g., an app on the user's device) that takes into account factors such as carbohydrate intake, the patient's insulin sensitivity factor (ISF), the patient's physiological condition (including current glucose levels), target glucose range, and indicates the number of units of insulin to be delivered. Too much insulin can cause hypoglycemia, while too little insulin can cause hyperglycemia.

[0033] For glycemic control in patients with type 1 diabetes, several recommended targets exist, such as a glucose management index (GMI) of less than approximately 7% (or approximately 53 mmol / mol), within the range of 70 mg / dL to 180 mg / dL for more than 70% of the time (TIR), less than 25% of the time above 180 mg / dL (TA180), less than 5% of the time above 250 mg / dL (TA250), less than 4% of the time below 70 mg / dL (TB70), and less than 1% of the time below 54 mg / dL (TB54). There is also growing interest in the newly proposed metric (the percentage of time within the tight glycemic range of 70 mg / dL to 140 mg / dL (TITR)) as a potential indicator of optimal CGM glycemic normality. However, there is currently no universally agreed-upon international TITR target in the management of type 1 diabetes.

[0034] Automated Insulin Delivery (AID) Systems (such as Medtronic MiniMed) TM The 780G glycemic index (GMI) has been shown to perform well in glycemic control, including the feasibility of achieving GMI targets close to or lower than 6.5%. However, in many glycemic control systems, some trade-offs may be necessary when achieving or exceeding these recommended targets, as efforts to improve one glycemic metric may have mixed effects on other CGM glycemic metrics. For example, increasing insulin dosage may decrease GMI, TA180, and TA250, but may potentially increase TB70 and TB54. Therefore, it may be desirable to determine whether meeting a specific glycemic target (e.g., a certain GMI target) corresponds to meeting certain targets (such as TIR, TITR, TB70, TB54, TA180, or TA250 targets) of other glycemic metrics, so that patients can expect appropriate targets and / or can use appropriate targets to control their glycemic control system (e.g., an AID system) to achieve the desired outcomes for a particular patient. Additionally, it may be desirable to explore dimensions of these CGM metrics to establish the meaning of targets, for example, over various time ranges, based on GMI targets.

[0035] According to certain embodiments, techniques are disclosed for determining corresponding targets for one or more CGM measures based on a target of one CGM measure. In one example, a classifier may be trained to classify CGM data samples, and the classification result of the classifier may be used to determine whether the performance of a blood glucose control system that meets a target of a first blood glucose measure (e.g., GMI) corresponds to the performance that meets a threshold target of a second blood glucose measure (e.g., TIR). For example, for a given target of the first blood glucose measure (e.g., GMI less than 7%, less than 6.8%, less than 6.6%, or less than 6.5%), classification results for different threshold target levels of the second blood glucose measure (e.g., TIR, TITR, TB70, TB54, TA180, or TA250) may be determined and used to generate a receiver operating characteristic (ROC) curve. Based on the ROC curve, an appropriate threshold target level for the second blood glucose measure that maximizes both the sensitivity and specificity of classification may be selected (e.g., the threshold target level associated with the point closest to the top left corner of the ROC curve). The selected threshold target level for the second blood glucose measure, which maximizes both the sensitivity and specificity of the classification, can be used as a target for the second blood glucose measure that can be associated with or correspond to the target of the first blood glucose measure. In this way, the corresponding target value and possible trade-offs between two or more different CGM measures can be determined and used to set patient expectations and control targets for the blood glucose control system. The target of the second blood glucose measure can be population-oriented or personalized.

[0036] Additionally, dimensionality or parameter reduction techniques (e.g., principal component analysis (PCA)) and correlation techniques (e.g., pairwise Pearson correlation) can be used to determine the association between GMI and various time measures across ranges based on CGM data samples, and to reduce the dimensionality of these CGM measures while preserving relevant information inherent in the CGM data samples (e.g., most of the variance). For example, time measures across ranges and GMI measures can be clustered to identify measures associated with GMI, such as those belonging to the same dimension or orthogonal to GMI, thereby validating the determination of targets within various time ranges associated with GMI targets.

[0037] The techniques disclosed herein can be implemented on servers, computers, user devices (e.g., smartphones), medical devices (e.g., CGM sensors or insulin delivery devices), etc. In one example, a user app implementing the techniques disclosed herein can execute on a user device to provide the user with corresponding targets for different CGM measures based on the target CGM measure input or selected by the user, enabling the user to understand the expected blood glucose control results and select appropriate targets that may be suitable for the user to configure the AID system.

[0038] In the following description, specific details are set forth for purposes of explanation in order to provide a thorough understanding of the examples of this disclosure. However, it will be apparent that various examples may be practiced without these specific details. For example, devices, systems, structures, components, methods, and other parts may be shown as parts in the form of block diagrams to avoid obscuring the examples with unnecessarily detailed information. In other instances, well-known devices, processes, systems, structures, operations, and techniques may be shown or may not be shown without the need for necessary detail to avoid obscuring the examples. The figures and descriptions are not intended to be limiting. The terms and expressions used in this disclosure are used as descriptive terms rather than limiting terms, and the use of these terms and expressions is not intended to exclude any equivalents of the features shown and described or portions thereof. The word “example” is used herein to mean “serves as an example, illustration, or description.” Any embodiment or design described herein as an “example” is not necessarily to be construed as preferred or advantageous over other embodiments or designs.

[0039] Carbohydrates ingested from food are broken down into glucose (sugar) in the stomach and / or intestines, and absorbed into the bloodstream in the small and / or large intestines. The bloodstream carries glucose to the body's capillaries, some of which diffuse into the interstitial fluid between cells (e.g., fat or muscle cells), where glucose can be used as energy. The human endocrine system (working in conjunction with the nervous system) directs and regulates bodily functions and activities, and may secrete chemicals that send messages to tissues and organs. For example, endocrine glands or organs may release hormones into the bloodstream for delivery to target cells with receptors. In one example, insulin can be produced by the beta cells of the pancreas and can, for example, increase glucose uptake, enhance glucose utilization, stop hepatic glucose production, stimulate glycogen formation in the liver and skeletal muscle, promote protein synthesis, and increase fat storage.

[0040] Insulin acts as a key that unlocks cells and helps glucose move into them, where it can be used as energy. Without insulin, glucose may not be able to enter cells for energy and could accumulate in the interstitial fluid and bloodstream. A healthy pancreas continuously releases small amounts of normal human insulin throughout the day, including between meals and during sleep. This small amount of insulin matches the amount of glucose released by the liver. A healthy pancreas may also secrete more insulin after eating to match the amount of food ingested.

[0041] The liver may absorb excess glucose during digestion and convert it into glycogen for storage, so that it can be released back into the bloodstream when needed. For example, when blood sugar levels are low, the alpha cells of the pancreas may secrete glucagon. Glucagon may cause the liver to release stored glucose (glycogen) into the bloodstream to help increase glucose levels. The balance between insulin secreted by beta cells and glucagon secreted by alpha cells helps maintain normal blood sugar levels, such as within the range of approximately 80 mg / dL to 140 mg / dL before meals.

[0042] Due to insufficient insulin production and / or insulin resistance, the insulin naturally produced in a diabetic patient's body may be insufficient to control glucose levels in the bloodstream. Therefore, diabetic patients may need to receive insulin from a pump or another delivery device (such as an injection or infusion device) to control glucose levels in their bloodstream. To control glucose levels, a typical therapeutic routine for diabetic patients generally includes a basal insulin dose and a bolus insulin dose. The dose of insulin to be delivered can be determined based on, for example, a carbohydrate count from a meal and / or the patient's glucose levels as measured using a glucose monitor (such as a continuous glucose monitor (CGM)).

[0043] In some implementations (e.g., in a closed-loop system), the insulin delivery device may communicate with or otherwise use sensor devices (including, but not limited to, CGMs) to perform various patient-specific measurements. In one example, the sensor device may include subcutaneously implanted electrodes to simultaneously monitor the patient's response to diet and insulin introduced by the insulin delivery device. The sensor device and the insulin delivery device may reside within a communication network (wired or wireless) with one or more processors and / or patient devices (such as a patient's smartphone equipped with an app or other software) to create a holistic system for monitoring the patient's disease state and facilitating their treatment.

[0044] Figure 1An example of a blood glucose level management system 100 according to certain embodiments is illustrated. The blood glucose level management system 100 can be used to monitor and regulate the blood glucose level of a patient 101. In the illustrated example, the blood glucose level management system 100 may include a delivery device 102, a monitoring device 104, a computing device 106, and an optional remote or cloud computing system 108. The delivery device 102, monitoring device 104, and computing device 106 may be embodied in various ways, including being housed in one or more device housings. For example, in some embodiments, all devices 102 to 106 may be housed in a single device housing. In some embodiments, each device 102 to 106 may be housed in a separate device housing. In some embodiments, two or more devices 102 to 106 may be housed in the same device housing. In some embodiments, a single device 102, 104, or 106 may have two or more portions housed in two or more housings. For example, the monitoring device 104 may include a body portion and a display and control portion that communicates with the body portion via wired or wireless means. Delivery device 102 may include body parts (e.g., including cannulas) and portions including reservoirs, pumps, and control units. These and other embodiments and combinations thereof are contemplated within the scope of this disclosure.

[0045] The blood glucose level management system 100 may include multiple communication links, such as communication links 112 to 118. Each of communication links 112 to 118 may be a wired connection and / or a wireless connection. In an embodiment where both devices are located in the same housing, the communication links may include, for example, wires, cables, and / or communication buses located on a printed circuit board. In an embodiment where the two devices are separated from each other in different device housings, the communication links may be wired connections and / or wireless connections. Wired connections may include, for example, Ethernet connections, Universal Serial Bus (USB) connections, and / or another type of physical connection. Wireless connections may include, for example, cellular connections, Wi-Fi connections, etc. Connections, mesh network connections, and / or another type of connection using wireless communication protocols. Some implementations of communication links 112 to 118 may use direct connections such as... Connections, and / or connections that can be routed through one or more networks or network devices (not shown), such as Ethernet networks, Wi-Fi networks, cellular networks, satellite networks, intranets, extranets, the Internet and / or other types of networks. Various combinations of wired and / or wireless connections may be used for communication links 112 to 118.

[0046] Delivery device 102 may be configured to deliver a therapeutic substance to patient 101. The therapeutic substance may include, for example, insulin, HIV medication, medication for treating pulmonary hypertension, iron chelating agents, analgesics, anticancer therapies, pharmaceuticals, vitamins, hormones, nutritional supplements, dyes, tracing media, saline media, hydration media, etc. Delivery device 102 may be attached to patient 101 (e.g., attached to patient 101's body or clothing) or may be at least partially implanted in patient 101's body. In some embodiments, delivery device 102 may include a reservoir, an actuator, a delivery mechanism, and a cannula (not shown). The reservoir may be configured to store a quantity of therapeutic substance. In some embodiments, the reservoir may be refillable or replaceable. The actuator may be configured to drive the delivery mechanism. In some examples, the actuator may include a motor, such as an electric motor. The delivery mechanism may be configured to move the therapeutic substance from the reservoir through the cannula. In some examples, the delivery mechanism may include a pump and / or a plunger. The cannula facilitates fluid connection between the reservoir and the patient 101's body. The cannula and / or needle facilitate the delivery of therapeutic substances to the patient 101's tissue layers, veins, interstitial fluid, or body cavities. During operation, an actuator may, in response to a signal (e.g., a command signal), drive the delivery mechanism, thereby moving the therapeutic substance from the reservoir through the cannula and into the patient 101's body.

[0047] The components of the delivery device 102 described above are provided by way of example only. The delivery device 102 may include other components, such as, but not limited to, a power supply, a communication transceiver, one or more processors or other computing resources, a memory device, and / or a user interface (e.g., buttons, keys, a display, etc.). In some embodiments, the delivery device 102 may host an app (e.g., an insulin calculator) that calculates the desired amount of therapeutic substance to be delivered to the patient 101. Those skilled in the art will recognize various embodiments of the delivery device 102 and the components of such embodiments. All such embodiments and components are contemplated within the scope of this disclosure.

[0048] The monitoring device 104 may be configured to detect the physiological condition of the patient 101 (e.g., glucose concentration level) and may also be configured to detect other physiological conditions. The monitoring device 104 may be attached to the body of the patient 101 (e.g., attached to the skin of the patient 101 via adhesive) and / or may be at least partially implanted in the body of the patient 101. Depending on the specific location or configuration, the monitoring device 104 may come into contact with the biological material of the patient 101 (e.g., interstitial fluid and / or blood).

[0049] Monitoring device 104 may include one or more sensors (not shown), such as, but not limited to, electrochemical sensors, electrical sensors, and / or optical sensors. As those skilled in the art will understand, electrochemical sensors may be configured to respond to the interaction or binding of a biomarker with an electrode by generating an electrical signal based on, for example, potential, conductance, current, and / or impedance along an electrical path through the electrode. The electrode may include a material selected to interact with a specific biomarker, such as glucose. Potential, current, conductance, and / or impedance may be correlated with the concentration of the specific biomarker. In one example, the electrochemical sensor may include a glucose limiting membrane (GLM) that limits the amount of glucose and oxygen delivered to a glucose oxidase (GOx) layer on the working electrode of the sensor to ensure that the reaction is glucose-limited. The GOx layer or another active enzyme layer on the working electrode of the sensor may decompose glucose and oxygen into gluconic acid and hydrogen peroxide. When a voltage signal is supplied to the working electrode, the generated peroxide molecules may interact with the working electrode to decompose hydrogen peroxide at the surface of the working electrode into two hydrogen ions, oxygen, and two electrons. Charge can be forced to move between electrodes (e.g., between the working electrode and the counter electrode), thereby generating a sensor current signal (Isig) that can be measured by sensor electronics. Other signals can also be measured, such as reverse voltage (Vcntr, the voltage potential difference between the counter electrode and the working electrode), electrochemical impedance spectroscopy (EIS) at different frequencies, etc. The signals measured using the sensor (including Isig, Vcntr, and EIS) can be processed (e.g., filtered or transformed) to generate other signals or parameters, such as a filtered Isig signal, real impedance and virtual impedance at various frequencies, etc. These signals and / or processed parameters can be used in one or more sensor glucose (SG) models (e.g., machine learning models or mathematical models) to determine SG values, which can be estimates of a patient's blood glucose (BG) level.

[0050] As those skilled in the art will understand, the electrical sensor can be configured to respond to an electrobiosignal by generating an electrical signal based on the amplitude, frequency, and / or phase of the electrobiosignal. The electrobiosignal may include changes in current resulting from the sum of potential differences across tissues (such as the nervous system) of patient 101. In some embodiments, the electrobiosignal may include a portion of the patient 101's heart over time that may indicate potential changes in the patient 101's glucose levels (e.g., recorded as an electrocardiogram). The optical sensor can be configured to respond to the interaction or binding of a biomarker to a substrate, for example, by generating an electrical signal based on changes in the brightness of a substrate. In one example, the substrate may include a material selected to fluoresce in response to contact with a selected biomarker (such as glucose). The fluorescence may be proportional to the concentration of the selected biomarker.

[0051] In some embodiments, the monitoring device 104 may include other types of sensors that can be worn, carried, or coupled to the patient 101 to measure activities of the patient 101 that may affect the patient 101's glucose levels or glycemic response. For example, the sensor may include an accelerometer configured to detect acceleration of the patient 101 or a part of the patient 101 (such as a person's hand or foot), whose positional changes may be correlated with the patient 101's activities. For example, acceleration or movement (or lack thereof) of a body part of the patient 101 may indicate the patient 101's exercise, sleep, or food / beverage consumption activities, which can affect the patient 101's glycemic response. In some embodiments, the sensor may measure heart rate and / or body temperature, which may indicate the amount of physical exercise experienced by the patient 101. In some embodiments, the sensor may include a Global Positioning System (GPS) receiver that detects GPS signals to determine the location of the patient 101.

[0052] The sensors described above are provided merely as examples. Other sensors or other types of sensors for monitoring physiological conditions, activities, and / or location will be recognized by those skilled in the art and are conceivable within the scope of this disclosure. For any sensor, the signal provided by the sensor may be referred to herein as a “sensor signal.” As used herein, the term “sensed data” may mean and include information represented by the sensor signal or by preprocessed sensor signals. In some embodiments, sensed data may include glucose levels in patient 101, acceleration of a portion of patient 101, heart rate of patient 101, temperature of patient 101, and / or geographic location of patient 101 (e.g., GPS location), etc. Monitoring device 104 may transmit sensed data to delivery device 102 via communication link 112 and / or to computing device 106 via communication link 114. The use of sensed data by delivery device 102 and / or computing device 106 is described in more detail below.

[0053] In some embodiments, monitoring device 104 may include components and / or circuitry configured to preprocess sensor signals. Preprocessing may include, for example, amplification, filtering, attenuation, scaling, isolation, normalization, transformation, sampling, and / or analog-to-digital conversion, etc. In some embodiments, monitoring device 104 may host an application for processing sensor signals. In some embodiments, monitoring device 104 may include a wired or wireless transceiver as described above for transmitting sensor signals or receiving commands or instructions. Those skilled in the art will recognize various specific implementations of such preprocessing, including but not limited to implementations using processors, controllers, integrated circuits, application-specific integrated circuits (ASICs), hardware, firmware, programmable logic devices, and / or machine-executable instructions, etc. The types of preprocessing and their specific implementations are provided merely as examples. Other types of preprocessing and specific implementations are contemplated within the scope of this disclosure. In some embodiments, monitoring device 104 may not perform preprocessing.

[0054] The computing device 106 can provide processing power and can be implemented in various ways. In some embodiments, the computing device 106 can be a consumer device such as a smartphone, a computerized wearable device (e.g., a smartwatch), a tablet computer, a laptop computer, or a desktop computer, etc., or it can be a dedicated device (e.g., a portable control device) provided by, for example, the manufacturer of the delivery device 102. In some embodiments, the computing device 106 can be a processing circuitry system that can be integrated with another device such as the delivery device 102. In some embodiments, the computing device 106 can be attached to the patient 101 (e.g., attached to the patient 101's body or clothing), can be at least partially implanted in the patient 101's body, and / or can be held in place by the patient 101.

[0055] For each embodiment of the computing device 106, the computing device 106 may include various types of logic circuit systems, including but not limited to microprocessors, controllers, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), central processing units (CPUs), graphics processing units (GPUs), programmable logic devices, memories (e.g., random access memory, volatile memory, non-volatile memory, etc.) or other discrete or integrated logic circuit systems, and combinations of such components. The term "processing circuit" may generally refer to any of the aforementioned logic circuits, alone or in combination with other logic circuits, or any other circuitry used to perform computations.

[0056] The delivery device 102, monitoring device 104, and computing device 106 have been described above. One or more of devices 102 to 106 may include a user interface (not shown) for presenting information to and / or receiving information from patient 101. The user interface may include a graphical user interface (GUI), a display device, a keyboard, a touchscreen, a speaker, a microphone, a vibration motor, buttons, switches, and / or other types of user interfaces. Those skilled in the art will recognize the various types of user interfaces that may be used, and all such user interfaces are conceivable within the scope of this disclosure. For example, in the case where computing device 106 is a consumer device such as a smartphone, tablet computer, laptop computer, etc., the user interface will include a display device, physical and / or virtual keyboard, and / or audio speakers, etc., provided by such a consumer device. In some embodiments, the user interface may notify patient 101 of sensed data (e.g., glucose levels) and / or insulin delivery data (e.g., historical, current, or future insulin delivery rates) and may present alerts to patient 101. In some embodiments, the user interface may receive input from patient 101, which may include, for example, requested changes to insulin delivery settings and / or meal instructions. The above description and implementation of the user interface are provided by way of example only, and other types and uses of the user interface are conceivable within the scope of this disclosure.

[0057] In a specific example, communication and collaboration between devices 102 and 106 can be used for insulin delivery. For example... Figure 1As depicted and as described above, devices 102 to 106 can communicate with each other via communication links 112 to 116. In some embodiments, computing device 106 can control the operation of delivery device 102 and / or monitoring device 104. For example, computing device 106 can generate one or more signals (e.g., command signals) that cause delivery device 102 to deliver insulin to patient 101, for example, at a basal dose and / or bolus dose. In some embodiments, computing device 106 can receive data associated with insulin delivery (e.g., insulin delivery data) from delivery device 102 and / or sensed data (e.g., glucose levels) from monitoring device 104, and can perform calculations to control delivery device 102 based on insulin delivery data, sensed data, and / or other data. Insulin delivery data may include, but is not limited to, the type of insulin delivered, historical insulin delivery rates and / or amounts, current insulin delivery rates and / or amounts, insulin delivery time, and / or user input affecting insulin delivery. As those skilled in the art will understand, in closed-loop operation mode, computing device 106 can communicate a dosing command to delivery device 102 based on, for example (received from monitoring device 104), the difference between the current glucose level in the patient 101's body and (for example, a target glucose level determined by computing device 106 or set on delivery device 102). The dosing command may indicate the amount of insulin to be delivered and / or the rate (or time) of insulin delivery, and may adjust the current glucose level toward the target glucose level.

[0058] The remote / cloud computing system 108 can be any proprietary remote / cloud computing system or commercial cloud computing system including one or more server computing devices. When the computing resources of a client computing device (e.g., computing device 106) are insufficient, the remote / cloud computing system 108 can provide alternative or additional computing resources as needed. Computing device 106 and the remote / cloud computing system 108 can communicate with each other via a communication link 118, which traverses one or more communication networks (not shown). Communication networks may include, for example, networks of the type such as Ethernet networks, Wi-Fi networks, cellular networks, satellite networks, intranets, extranets, the Internet, and / or Internet backbones. Those skilled in the art will recognize specific implementations of the remote / cloud computing system 108 and how to interact with such a system via various types of networks. For example, the remote / cloud computing system 108 may include an array of processing circuitry systems and be capable of executing machine-readable instructions. Such specific implementations, interfaces, and networks are contemplated within the scope of this disclosure.

[0059] In some embodiments, the remote / cloud computing system 108 may make a treatment determination (e.g., insulin dosage or adjusted insulin dosage) and may communicate the treatment to the delivery device 102 via computing device 106. In some embodiments, computing device 106 may make a treatment determination and communicate that determination to delivery device 102. In some embodiments, monitoring device 104 may make a treatment determination and communicate that determination directly or through an intermediary such as computing device 106 to delivery device 102.

[0060] Figure 2 This is a block diagram of an example of a blood glucose level management system 200 according to certain embodiments. In the illustrated example, the blood glucose level management system 200 may include a glucose sensor subsystem 210, a controller 220, an insulin delivery subsystem 230, a glucose delivery subsystem 240, and a glucagon delivery subsystem 250. The glucose sensor subsystem 210 may generate sensor glucose (SG) signals (e.g., SG levels) (these SG signals may be estimates of blood glucose levels in body 260) and may provide the SG signals to the controller 220. The controller 220 may receive the SG signals and generate commands to the insulin delivery subsystem 230, and in some embodiments, generate commands to the glucose delivery subsystem 240 and / or the glucagon delivery subsystem 250. The insulin delivery subsystem 230 may receive commands from the controller 220 and deliver insulin to body 260 according to the commands. In some embodiments, the glucose delivery subsystem 240 may receive commands from the controller 220 and provide glucose to body 260 according to the commands. In some implementations, the glucagon delivery subsystem 250 may receive a command from the controller 220 and deliver glucagon into the body 260 according to the command.

[0061] In some implementations, the glucose sensor subsystem 210 may include a glucose sensor, sensor electronics configured to generate an SG signal, a sensor communication system configured to transmit the SG signal to a controller 220, and a housing for the sensor electronics and the sensor communication system. The glucose sensor can measure blood glucose levels, for example, directly from the bloodstream, or indirectly via interstitial fluid using a subcutaneous sensor, as described in more detail below.

[0062] Controller 220 may include electrical components and software for generating commands to insulin delivery subsystem 230, glucose delivery subsystem 240, and / or glucagon delivery subsystem 250. Controller 220 may include a controller communication system to receive sensor signals and provide commands to insulin delivery subsystem 230, glucose delivery subsystem 240, and / or glucagon delivery subsystem 250. In some embodiments, controller 220 may implement a glucose calculator. In some embodiments, controller 220 may include a user interface and / or operator interface (not shown), which includes data input devices and / or data output devices. Such data output devices may, for example, generate signals to initiate alarms and / or include a display or printer to display the status of controller 220 and / or the patient's vital signs. Such data input devices may include dials, buttons, pointing devices, manual switches, alphanumeric keys, touch-sensitive displays, combinations thereof, etc., for receiving user input and / or operator input. For example, such data input devices may be used to schedule and / or initiate insulin bolus injections for meals. However, it should be understood that these are merely examples of input and output devices that may be part of the operator and / or user interface, and the subject matter for which protection is sought is not limited in these respects.

[0063] Insulin delivery subsystem 230 may include, for example, an infusion device and / or infusion tubing to infuse insulin into body 260. Similarly, glucose delivery subsystem 240 may include, for example, an infusion device and / or infusion tubing to infuse glucose into body 260. Similarly, glucagon delivery subsystem 250 may include, for example, an infusion device and / or infusion tubing to infuse glucagon into body 260. In some embodiments, a shared delivery system and / or infusion tubing may be used to infuse insulin, glucagon, and / or glucose into body 260. In some embodiments, an intravenous system for providing fluids to a patient (e.g., in a hospital or other medical setting) may be used to infuse insulin, glucagon, and / or glucose. However, it should be understood that some exemplary embodiments may include insulin delivery subsystem 230 without glucagon delivery subsystem 250 and / or without glucose delivery subsystem 240. In some embodiments, each of the insulin delivery subsystem 230, glucose delivery subsystem 240, and glucagon delivery subsystem 250 may include infusion electrical components that activate an infusion motor according to a command from controller 220, an infusion communication system that receives commands from controller 220, and a delivery subsystem housing.

[0064] In some embodiments, controller 220 may be housed within a delivery subsystem housing, and the infusion communication system may include electrical traces or wires carrying commands from controller 220 to the delivery subsystem. In some embodiments, controller 220 may be housed within a sensor system housing, and the sensor communication system may include electrical traces or wires carrying sensor signals from sensor electrical components to controller electrical components. In some embodiments, controller 220 may have its own housing or may be included in an auxiliary device. In some embodiments, controller 220 may share a single housing with the delivery subsystem and sensor system. In some embodiments, the sensor, controller, and / or infusion communication system may utilize cables; wires; fiber optic cables; RF, IR, or ultrasonic transmitters and receivers; combinations thereof; and / or the like, in place of electrical traces, to name just a few.

[0065] In some implementations, the blood glucose level management system 200 may further include a dietary intake monitoring subsystem 215. The dietary intake monitoring subsystem 215 may be used to record or automatically detect a user's food intake. For example, in some implementations, the user may input the food items consumed at the time of the meal and / or an estimated amount of carbohydrates in the meal. In some implementations, the dietary intake monitoring subsystem 215 may include sensors (e.g., a camera or accelerometer) that automatically detect meal events, the food items consumed at the time of the meal, and / or an estimated amount of carbohydrates in the meal. The estimated amount of carbohydrates in the meal may be transmitted to a controller 220, which may determine the appropriate amount of the dietary bolus and generate a command for the insulin delivery subsystem 230 to deliver the dietary bolus. In some implementations, the dietary intake monitoring subsystem 215 may not be used in the blood glucose level management system 200 and may determine the dosage of the dietary bolus based on measured glucose levels.

[0066] Figure 3 This is a perspective view of an example CGM system 300 according to certain embodiments. In the illustrated example, the CGM system 300 may include a sensor kit 302 provided for subcutaneous placement of an active portion, such as a flexible sensor 310, at a selected site within a user's body. The subcutaneous or percutaneous portion of the sensor kit 302 includes a hollow slotted insertion needle 326 with a sharp tip 312 and a cannula 322. The sensor kit 302 facilitates accurate placement of the flexible sensor 310 into the user's body. The glucose sensing portion 318 of the flexible sensor 310 is located within the cannula 322. The glucose sensing portion 318 includes one or more sensor electrodes 316, which may be exposed to the user's bodily fluids 390, for example, through a window 314 formed in the cannula 322. The sensor electrodes 316 may include, for example, a counter electrode, a reference electrode, and one or more working electrodes.

[0067] The proximal portion of the flexible sensor 310 can be mounted in a mounting base 305 suitable for placement on a user's skin. In some embodiments, the mounting base 305 may be a pad having a lower surface coated with a suitable pressure-sensitive adhesive layer 334, wherein a release liner 336 is typically provided to cover and protect the adhesive layer 334 until the sensor kit 302 is ready for use. The mounting base 305 may include an upper layer 330 and a lower layer 331, wherein a connection portion 328 of the flexible sensor 310 is sandwiched between the upper layer 330 and the lower layer 331. The connection portion 328 may include a front segment of a glucose sensing portion 318 of the flexible sensor 310, which may be folded at an angle to extend downward through a hole 320 formed in the lower layer 331 of the mounting base 305. Optionally, the adhesive layer 334 (or another portion of the device in contact with in vivo tissue) may include an anti-inflammatory agent for reducing inflammatory responses and / or an antibacterial agent for reducing the chance of infection. The insertion needle 326 may be adapted to be received via a needle port 324 formed in the upper layer 330 of the mounting base 305 and via a lower hole 320 in the lower layer 331 of the mounting base 305. In some embodiments, the insertion needle 326 may be withdrawn after insertion to leave the glucose sensing portion 318 (including sensor electrode 316) and / or the cannula 322 in the appropriate position at the selected insertion site.

[0068] The flexible sensor 310 of the sensor kit 302 can be connected to sensor electronics 304, which may be referred to as a transmitter. For example, the mounting base 305 may be designed such that a glucose sensing portion 318 can be coupled to a connection portion 328, which terminates at a conductive contact pad, etc. The connection portion 328 and the contact pad may be adapted to be electrically connected to the sensor electronics 304 to determine a user's glucose level in response to a signal from the sensor electrode 316. The connection portion 328 may be electrically connected to the sensor electronics 304 via a connector block 338. The sensor electronics 304 may include, for example, a housing 350 supporting a printed circuit board 344, a battery 354, and an antenna 346. In some embodiments, the housing 350 may include an upper housing 356 and a lower housing 352, which are sealed, for example, by ultrasonic welding to form a waterproof (or water-resistant) seal, thereby allowing cleaning by immersion (or wiping) with water, detergents, alcohol, etc. In some embodiments, the upper housing 356 and the lower housing 352 may be formed of medical-grade plastic. In alternative embodiments, the upper housing 356 and lower housing 352 may be joined together and bonded together by other methods (such as snap-fit, sealing rings, RTV (siloxane sealant)), or formed from other materials such as metal, composite materials, ceramics, etc. In other embodiments, the components may be encapsulated in epoxy resin or other moldable materials that are compatible with the electronics and reasonably moisture-proof. In some embodiments, sensor electronics 304 may be connected to the flexible sensor 310 via connector block 338 and connector 340, and may be mounted on mounting base 305. In some embodiments, sensor electronics 304 may be connected to the flexible sensor 310 via connector block 338, connector 340, and cable 342. In some embodiments, the bottom surface of the lower housing 352 may be coated with a suitable pressure-sensitive adhesive layer 358, wherein a release liner 348 is typically provided to cover and protect the adhesive layer 358 until the sensor electronics 304 is ready for use.

[0069] Battery 354 may include a rechargeable or non-rechargeable battery. In some embodiments, battery 354 may include a silver oxide battery cell. In other examples, different battery chemistry, such as lithium-based chemistry, alkaline batteries, nickel metal hydride, etc., may be used, and different numbers of batteries may be used. Battery 354 may power sensor kit 302 via sensor electronics 304 through, for example, connector block 338, connector 340, and / or cable 342.

[0070] Sensor electronics 304 may include a processor, transceiver unit, signal processing / conditioning circuitry, and memory device on a printed circuit board 344. The processor may be configured to receive current and / or voltage signals from sensor electrodes 316 of the flexible sensor 310. Sensor electrodes 316 may generate a sensor signal indicating the concentration of the analyte being measured (e.g., glucose). In one example, the sensor signal may include a current measured at a working electrode, which may include glucose oxidase (GOx) to catalyze the reaction with glucose. The sensor signal may also include a voltage signal measured at the counter electrode of sensor electrode 316 or between the working electrode and the counter electrode (or reference electrode). In some examples, sensor electrodes 316 inserted into a user's body may also be used to measure other electrical parameters, such as capacitance, resistance, impedance, etc.

[0071] Examples of impedance parameters measured using sensor electrode 316 may include electrochemical impedance spectroscopy (EIS) values. EIS can provide additional information in the form of sensor impedance and impedance-related parameters at different frequencies. Furthermore, for certain frequency ranges, impedance and / or impedance-related data can be substantially glucose-independent. This glucose independence allows for the generation of robust, highly reliable sensor glucose values ​​using a variety of EIS-based markers or indices (e.g., through fusion methods), and also enables the assessment of the condition, health, age, and efficiency of individual electrodes and the overall sensor, substantially independent of glucose-related current signals. In one example, analysis of glucose-independent impedance data can provide information about the efficiency of the glucose sensor, i.e., how quickly the glucose sensor hydrates and prepares for data acquisition. Additionally, glucose-independent impedance data (e.g., using values ​​of actual impedance at 1 kHz) can provide information about potential blockages that may exist on the sensor membrane surface, which could temporarily prevent glucose from entering the sensor and thus cause a signal drop. Furthermore, glucose-independent impedance data, for example, using phase angle and / or virtual impedance values ​​at 1 kHz and higher frequencies, can provide information about sensor sensitivity loss during long-term wear.

[0072] The processor of sensor electronics 304 can receive sensor signals and calibrate them using a calibration function, which may be determined, for example, using reference values. In some embodiments, the parameters of the calibration function (e.g., calibration factor or sensor sensitivity) may be determined (e.g., using calibration data) or predetermined and stored in memory for the processor's use. For example, in some embodiments, the processor may implement or otherwise execute a calibration application module that calculates or otherwise determines calibration parameters based on measurements and reference values ​​(e.g., blood glucose values ​​measured using, for example, a finger prick blood glucose meter). Based on the sensor signals and the determined calibration function, the processor can estimate the user's blood glucose level. In some embodiments, the processor may store glucose measurements in memory. Sensor measurement results may be displayed on a display of sensor electronics 304 or transmitted to receiver 306 for display.

[0073] The memory of sensor electronics 304 can be any type of memory device and can be configured to store raw sensor signals from flexible sensor 310 and / or glucose measurements (e.g., SG values) generated by the processor, calibration parameters for determining glucose measurements from sensor signals, or other data used and / or generated by the processor. The memory may also store software and / or firmware executable by the processor. In some embodiments, the memory may also include configuration information for configuring the operation of sensor electronics 304.

[0074] The signal processing / conditioning circuit of the sensor electronics 304 may include one or more amplifiers, filters, current-to-voltage converters, integrators, analog-to-digital converters, potentiometers, etc. The signal processing / conditioning circuit may filter signals, apply a constant reference voltage level, amplify current or voltage signals, sample analog current or voltage signals, and convert the analog current or voltage signals into digital data values, etc.

[0075] The transceiver unit of sensor electronics 304 may include a wireless transceiver that can transmit glucose measurements determined by a processor and stored in memory to receiver 306 using a suitable wireless communication protocol (such as Bluetooth (e.g., Bluetooth Low Energy (BLE)), Wi-Fi, WiMax, etc.). For example, the transceiver unit may transmit radio signals indicating glucose levels at regular time intervals (e.g., every 5 minutes) to provide real-time sensor glucose (SG) values. The transceiver unit may also receive data from receiver 306, such as configuration data and calibration data (e.g., reference glucose values). In some embodiments, the transceiver unit of sensor electronics 304 may transmit data to receiver 306 via cable or wire.

[0076] Receiver 306 may include a monitor 335 for monitoring physiological characteristic readings and / or generating warning signals, and a display 332 for displaying physiological characteristic readings and / or warning signals. For example, SG values / charts may be displayed on the display 332 of receiver 306, allowing a user to monitor blood glucose levels and administer insulin using an insulin pump. In various embodiments, receiver 306 may be implemented in a desktop computer, pager, television with communication capabilities, laptop computer, server, network computer, personal digital assistant (PDA), smartphone, infusion pump with display, glucose sensor with display, combined infusion pump / glucose sensor, etc.

[0077] Figure 4A and Figure 4B An example of an integrated glucose sensor device 400 according to certain embodiments is depicted, which may implement some of the examples disclosed herein. The glucose sensor device 400 may be an example of a monitoring device 104, a glucose sensor subsystem 210, or a CGM system 300. In the illustrated example, the glucose sensor device 400 may include a housing formed by a substrate 410 and a top cover 420, sensor electronics encapsulated within the housing, an electrochemical sensor 440 partially located within the housing and connected to the sensor electronics, and an adhesive patch 430 attached to the substrate 410. The electrochemical sensor 440 may be an example of the aforementioned electrochemical sensor and may include multiple electrodes located at a distal end that can be inserted into the user's subcutaneous layer. The proximal end of the electrochemical sensor 440 is flexible and held against a printed circuit board (PCB) 412 by an elastomer or rubber block 414, such that contacts on the proximal end of the electrochemical sensor 440 are physically and electrically connected to contact pads on the PCB 412. The housing may be hermetically sealed to protect the sensor electronics from moisture. The electrochemical sensor 440 can be inserted into the subcutaneous layer of a user using an inserter. The inserter may include a needle and may encapsulate the entire glucose sensor device 400 before insertion. When activated (e.g., pushed), the inserter may move the entire glucose sensor device 400 toward the user's skin to attach the adhesive patch 430 to the user's skin and insert the distal end of the electrochemical sensor 440 into the user's subcutaneous layer using the needle, wherein the needle may be automatically retracted, for example, after the distal end of the electrochemical sensor 440 has been inserted into the user's body.

[0078] The sensor electronics within the housing may include battery 416, antenna 418, and circuitry on PCB 412. The circuitry on PCB 412 may include, for example, a processor, controller, potentiometer, filter, amplifier, wireless (e.g., Bluetooth) transceiver, one or more memory devices, power conversion and management devices, or combinations thereof. The predicted in vivo characteristics of the electrochemical sensor 440 and the SG model described above may be stored in one or more memory devices. Before or after insertion of the distal end of the electrochemical sensor 440 into the user's subcutaneous layer, the sensor electronics may be activated to initialize the sensor for measuring sensor measurement data (e.g., Isig, Vcntr, EIS, etc.). The processor or controller may execute instruction code stored on one or more memory devices to obtain the sensor measurement data, process (e.g., filter, amplify, transform, etc.) the sensor measurement data, estimate SG values ​​using the SG model and the sensor measurement data, optionally clear certain estimated SG values, and transmit the estimated SG values ​​to a receiver (e.g., a smartphone or insulin pump) via the wireless transceiver and antenna 418, as described above.

[0079] Figure 5 An example of a delivery device 500 is depicted, which can implement the aforementioned delivery device (e.g., delivery device 102 or insulin delivery subsystem 230). In an insulin delivery device implemented using delivery device 500, insulin delivery can be performed based on internal communication between a central computing module (e.g., the microprocessor of delivery device 500) and an insulin delivery module (e.g., including a microcontroller, motor, and pump). For example, insulin delivery can be initiated by the central computing module transmitting a delivery command in the form of an electrical signal traveling to the insulin delivery module via a communication structure. The central computing module can also be configured to communicatively couple to a remote or cloud computing system (e.g., Figure 1 The computing devices of the remote / cloud computing system 108 (e.g., Figure 1 The delivery device 500 communicates with a computing device 106 (e.g., via a transceiver). According to the process described herein, the delivery device 500 can transmit various event data to a remote or cloud computing system, which can then transmit insulin delivery confirmations to the delivery device 500.

[0080] The delivery device 500 can deliver insulin via a tubing 510 configured for fluid connection to a subcutaneously inserted cannula. The delivery device 500 can be configured to deliver two types of doses: a basal dose, which can be delivered periodically (e.g., every five minutes) throughout the day and night in micro-dose doses; and a bolus dose to cover and / or correct hyperglycemia caused by meals. In the depicted example, the delivery device 500 includes a user interface with button elements 520 manipulating insulin bolus administration, changing therapy settings, altering user preferences, and selecting display features, etc. The delivery device 500 may also include a display device 530 for presenting various types of information or data to the user. According to various aspects of this disclosure, a user of the delivery device 500 can use the button elements 520 to input certain event data (e.g., event type, event start time, event details, etc.) and can use the display device 530 to confirm the user input. Figure 5 The delivery device 500 is provided by way of example only, and other types of insulin delivery devices and other processes different from those described above are considered to be within the scope of this disclosure.

[0081] Based on glucose levels measured using a CGM (e.g., CGM system 300 or glucose sensor device 400), various metrics can be used to determine the performance of a blood glucose control system (such as a human or AID system, including delivery device 500). For example, the Glucose Management Index (GMI) can be used to describe the average (mean) glucose level measured using a CGM over a period of time (e.g., approximately 10 to 14 days). GMI can be determined based on the following:

[0082] GMI (percentage) = 3.31 + 0.02392 × (glucose reading, in mg / dL).

[0083] A GMI below approximately 5.7% indicates a normal condition, a GMI between approximately 5.7% and approximately 6.5% indicates a prediabetes condition, and a GMI greater than approximately 6.5% indicates a diabetes condition. For most adults with type 1 diabetes, the goal is likely to maintain a GMI below approximately 7% (or approximately 154 mg / dL). Time-range measures can be used to describe the frequency of a person's glucose levels across different ranges.

[0084] Figures 6A to 6C An example of continuous glucose monitoring results is shown. Figure 6AIn the example shown, curve 610 illustrates glucose levels measured over 24 hours, and pie chart 612 presents statistics on glucose levels measured over 24 hours. Curve 610 and pie chart 612 show that glucose levels measured in the illustrated example range from 70 mg / dL to 180 mg / dL for 100% of the day. Figure 6B In the example shown, curve 620 illustrates glucose levels measured over 24 hours, and pie chart 622 presents statistics on glucose levels measured over 24 hours. Curve 620 and pie chart 622 show that, over 24 hours, glucose levels measured in the illustrated example ranged from 70 mg / dL to 180 mg / dL for approximately 70% of the time, were below 70 mg / dL for approximately 5% of the time, and were above 180 mg / dL for approximately 25% of the time. Figure 6C In the example shown, curve 630 illustrates glucose levels measured over 24 hours, and pie chart 632 illustrates the statistics of glucose levels measured over 24 hours. Curve 630 and pie chart 632 show that, over 24 hours, the glucose levels measured in the illustrated example were in the range of 70 mg / dL to 180 mg / dL for approximately 40% of the time, below 70 mg / dL for approximately 20% of the time, and above 180 mg / dL for approximately 40% of the time.

[0085] Figure 7 Examples of recommended targets for range-of-time metrics used to assess blood glucose control systems are illustrated. Figure 7 In the example shown, recommended TIR targets for glycemic control in patients with type 1 diabetes may include: more than 70% of the time within the range of 70 mg / dL to 180 mg / dL (or approximately 3.9 mmol / L to 10.0 mmol / L) (TIR), less than 25% of the time above 180 mg / dL (or 10.0 mmol / L) (TA180), less than 5% of the time above 250 mg / dL (or 13.9 mmol / L) (TA250), less than 4% of the time below 70 mg / dL (or approximately 3.9 mmol / L) (TB70), and less than 1% of the time below 54 mg / dL (or 3.0 mmol / L) (TB54). There is also growing interest in the newly proposed metric (percentage of time within the tight glycemic range of 70 mg / dL to 140 mg / dL (TITR)) as a potential indicator of optimal CGM glycemic normality.

[0086] Automated Insulin Delivery (AID) Systems (such as Medtronic MiniMed) TMThe 780G glycemic index (GMI) has been shown to perform well in glycemic control, including the feasibility of achieving GMI targets close to or lower than 6.5%. However, in many glycemic control systems, some trade-offs may be necessary when achieving or exceeding these recommended targets, as efforts to improve one glycemic metric may have mixed effects on other CGM glycemic metrics. For example, increasing insulin dosage may decrease GMI, TA180, and TA250, but may potentially increase TB70 and TB54. Therefore, it may be desirable to determine whether meeting a specific glycemic target (e.g., a certain GMI target) corresponds to meeting certain targets (such as TIR, TITR, TB70, TB54, TA180, or TA250 targets) of other glycemic metrics, so that patients can expect appropriate targets and / or can use appropriate targets to control their glycemic control system (e.g., an AID system) to achieve the desired outcomes for a particular patient. Additionally, it may be desirable to explore dimensions of these CGM metrics to establish the meaning of targets, for example, over various time ranges, based on GMI targets.

[0087] According to certain embodiments, techniques are disclosed for determining corresponding targets for one or more CGM measures based on a target of one CGM measure. In one example, a classifier may be trained to classify CGM data samples, and the classification result of the classifier may be used to determine whether the performance of a blood glucose control system that meets a target of a first blood glucose measure (e.g., GMI) corresponds to the performance that meets a threshold target of a second blood glucose measure (e.g., TIR). For example, for a given target of the first blood glucose measure (e.g., GMI less than 7%, less than 6.8%, less than 6.6%, or less than 6.5%), classification results for different threshold target levels of the second blood glucose measure (e.g., TIR, TITR, TB70, TB54, TA180, or TA250) may be determined and used to generate a receiver operating characteristic (ROC) curve. Based on the ROC curve, an appropriate threshold target level for the second blood glucose measure that maximizes both the sensitivity and specificity of classification may be selected (e.g., the threshold target level associated with the point closest to the top left corner of the ROC curve). The selected threshold target level for the second blood glucose measure, which maximizes both the sensitivity and specificity of the classification, can be used as a target for the second blood glucose measure that can be associated with or correspond to the target of the first blood glucose measure. In this way, the corresponding target value and possible trade-offs between two or more different CGM measures can be determined and used to set patient expectations and control targets for the blood glucose control system. The target of the second blood glucose measure can be population-oriented or personalized.

[0088] Additionally, dimensionality or parameter reduction techniques (e.g., principal component analysis (PCA)) and correlation techniques (e.g., pairwise Pearson correlation) can be used to determine the association between GMI and TIR measures based on CGM data samples, and to reduce the dimensionality of these CGM measures while preserving relevant information inherent in the CGM data samples (e.g., most of the variance). For example, TIR measures and GMI measures can be clustered to identify measures associated with GMI, such as those belonging to the same dimension or orthogonal to GMI, thereby validating the determination of targets over various timeframes associated with GMI targets.

[0089] In one example, samples were collected from patients with type 1 diabetes and treated with Medtronic MiniMed. TM Real-world data from individuals treated with the 780G system were used, and artificial intelligence and machine learning techniques were employed to determine appropriate targets for TIR, TITR, TB70, TB54, TA180, and TA250, which correspond to GMI targets of less than approximately 7%, less than approximately 6.8%, less than approximately 6.6%, and less than 6.5%, respectively. For example, the data could be used to predict whether achieving a GMI target of, for example, less than 7% or another value would guarantee compliance with consensus targets for TIR, TB70, TB54, TA180, and TA250. The data was also analyzed to identify and quantify trade-offs in glycemic control when pursuing stricter GMI benchmarks. The data was also used to determine the associations and dimensions of these CGM measures and to establish the meaning of CGM targets based on GMI targets. Results showed that targets for hypoglycemic time (e.g., TB70 and TB54) may not be reliably determined based on GMI targets, but targets for TB54 can be reliably determined based on TB70 targets. The results can be used to make decisions about treatment goals and to set achievable CGM glucose measures, which can be used to set realistic expectations for diabetes management using the AID system.

[0090] The data used in this example includes data from Medtronic MiniMed over a period longer than two years. TM 780G system uploaded to Medtronic CareLink TM Anonymous real-world data, including data from 14,459 individuals aged 15 or younger and 41,692 individuals aged 15 or older. Medtronic MiniMed TMThe 780G system can operate in open-loop mode or advanced hybrid closed-loop (AHCL or automatic) mode, where current and past glucose levels, as well as trends in glucose levels, measured by a CGM sensor, can be used to predict, adjust, and correct insulin delivery. The 56,151 individuals included users from various countries. Each of the 56,151 individuals had type 1 diabetes (TID) and was registered with CareLink. TM Individual accounts have been created and data has been agreed to be used for research. Each individual has 10 or more days of sensor glucose (SG) data after initial AHCL activation. All available CGM data from all qualified users (analysis cohorts) are used, regardless of whether the system is in AHCL or open-loop mode.

[0091] CGM data can be partitioned into multiple CGM data samples. For example, each CGM data sample may include CGM data of a user over a period of time (e.g., approximately 14 days to 10 days or longer). The GMI and TIR metrics of multiple CGM data samples can be determined and used as training and / or test samples. To avoid overfitting in the identification of appropriate targets and to perform unbiased evaluation, CGM data can be partitioned into a training dataset (e.g., random CGM data samples from approximately 70% of users) and a test dataset (e.g., CGM data samples from the remaining 30% of users). In one example, the training dataset can be used to determine appropriate CGM targets, and the test dataset can be used to independently evaluate the classifier and the sensitivity, specificity, accuracy, and other performance parameters of the determined targets. In some examples, bootstrap samples from the training dataset can be used to compute the confidence interval and area under the ROC curve (AUC) of the determined targets.

[0092] In this example, among the 14,459 individuals aged 15 or younger, 46.8%, 53.0%, and 0.2% reported themselves as male, female, and private, respectively. Among the 41,692 individuals aged 15 or older, 42.8%, 57.0%, and 0.2% reported themselves as male, female, and private, respectively. The data from the 14,459 individuals aged 15 or younger were split into a training dataset (containing data from 10,121 users) and a test dataset (containing data from 4,338 users). These 14,459 users used Medtronic MiniMed. TMThe median duration of the 780G system was approximately 219 days (96 days at the 25th percentile and 391 days at the 75th percentile). The 41,692 individuals aged 15 and older comprised 27.6% of users aged 16–28, 29.9% aged 29–42, 25.5% aged 43–55, and 16.9% aged 56 and older. The data from these 41,692 users was split into a training dataset (containing data from 29,184 users) and a test dataset (containing data from 12,508 users). The 41,692 users aged 15 and older used Medtronic MiniMed. TM The median duration of the 780G system was 179 days (with 77 days at the 25th percentile and 335 days at the 75th percentile).

[0093] Figure 8 The above example of a CGM dataset is shown in some implementation scheme for determining the target of a CGM metric based on the target of another CGM metric. Figure 8 The data presented include total time statistics under AHCL mode for a group of 14,459 individuals aged 15 years or younger and a group of 41,692 users aged 15 years or older, sensor glucose (SG) values ​​(including mean and standard deviation (mean STD)), the percentage of users achieving each glucose target among various glucose targets (e.g., GMI < 7%, TIR > 70%, TITR > 50%, TB70 < 4%, TB54 < 1%, TA180 < 25%, and TA250 < 5%), and the average percentage of time within each of several SG ranges.

[0094] Figure 8 A similar summary of user data is also shown, with users consistently adhering to the recommended settings for maintaining a glucose target of 100 mg / dL and an active insulin time (AIT) of 2 hours for more than 95% of the time. For example, for users aged 15 years or younger who consistently used the recommended settings, the mean STD values ​​for TIR, TITR, TB70, TB54, TA180, and TA250 measures were 76.9% to 7.5%, 55.4% to 8.4%, 2.9% to 2.0%, 0.6% to 0.6%, 20.1% to 7.7%, and 4.2% to 3.3%, respectively. For users older than 15 years who consistently use the recommended settings, the mean STD values ​​for the TIR metric, TITR metric, TB70 metric, TB54 metric, TA180 metric, and TA250 metric are 80.07.4%, 56.49.0%, 2.21.7%, 0.40.5%, 17.87.7%, and 3.02.7%, respectively.

[0095] Figure 9 Another summary is shown of an example of a CGM dataset used to determine the target of a CGM metric based on the target of another CGM metric, according to certain implementation schemes. Figure 9 The data presented include total time statistics, sensor glucose (SG) values ​​(including mean and standard deviation), mean GMI, and average percentage of time in each of several SG ranges for a group of 14,459 individuals aged 15 years or younger and a group of 41,692 users aged 15 years or older under AHCL mode. Figure 9 The mean STD value of GMI for all users is shown to be approximately 6.90.4.

[0096] Figure 9 A similar summary of user data is also shown, showing that users consistently adhere to recommended settings for maintaining a glucose target of 100 mg / dL and an active insulin time (AIT) of 2 hours for more than 95% of the time. Figure 9 The figure shows that for users who consistently adhere to the recommended settings, the mean STD value of GMI can be approximately 6.70.3.

[0097] To predict the appropriate targets associated with GMI targets of less than 7%, less than 6.8%, less than 6.6%, and less than 6.5% for TIR, TITR, TB70, TB54, TA180, and TA250, the aforementioned training samples can be used to train multiple classifiers to predict whether CGM data samples that meet a specific GMI target (e.g., 7%, 6.8%, 6.6%, or 6.5%) also meet the target for the time-in-range metric. The multiple classifiers can include binary classifiers, such as logistic regression models, support vector machines, neural networks, decision trees, etc. A binary classifier can classify each CGM data sample that meets a specific GMI target into one of two groups, such as having a time-in-range metric value above or below a threshold. For example, one of the multiple classifiers can be used to classify a CGM data sample with a GMI less than 7% as having a TIR greater than a certain threshold (e.g., a positive result) or having a TIR less than a threshold (e.g., a negative result). One classifier among multiple classifiers can be used to classify CGM data samples with a GMI of less than 7% as having a TA180 below a certain threshold (e.g., a positive result) or having a TA180 above that threshold (e.g., a negative result). Yet another classifier among multiple classifiers can be used to classify CGM data samples with a GMI of less than 6.8% as having a TIR above a certain threshold (e.g., a positive result) or having a TIR below that threshold (e.g., a negative result).

[0098] The classification result of a classifier using a threshold can be compared with a known TIR metric of the CGM data samples to determine whether the classification result is a true positive (TP), false positive (FP), true negative (TN), or false negative (FN). In this way, the performance of a classifier can be determined using training samples against a specific threshold, such as the classifier's accuracy, true positive rate, true negative rate, false positive rate, and false negative rate. Accuracy can be the ratio between the total number of correctly predicted predictions and the total number of predictions: Accuracy = (TP + TN) / (TP + FP + TN + FN). The true positive rate (TPR, sensitivity, or recall) can be the ratio between the number of true positive predictions and the total number of true positive and false negative predictions (e.g., TPR = TP / (TP + FN)). The false positive rate (FPR) can be the ratio between the number of false positive predictions and the total number of false positive and true negative predictions (e.g., FPR = FP / (FP + TN)). The true negative rate (TNR, or specificity) can be the ratio between the number of true negative predictions and the total number of true negative and false positive predictions (e.g., TNR = TN / (TN + FP)). The false negative rate (FNR) can be the ratio between the number of false negative predictions and the total number of false negative and true positive predictions (e.g., FNR = FN / (FN + TP)). For each classifier, the true positive rate and false positive rate for each of a plurality of thresholds can be determined as described above, and the true positive rate and the corresponding false positive rate for the plurality of thresholds can be plotted in a receiver operating characteristic (ROC) curve.

[0099] Figure 10 Graph 1000, including an example of a receiver operating characteristic (ROC) curve 1020, illustrates an example of the performance of a classifier using various classification thresholds to determine the target of one CGM metric based on another. The horizontal axis of Graph 1000 corresponds to the false positive rate, and the vertical axis corresponds to the true positive rate. The true positive rate (sensitivity or recall) represents the proportion of samples predicted as positive when they are actually positive. The false positive rate represents the proportion of samples predicted as positive when they are actually negative. Each ROC curve is a graph showing the performance of a binary classification model at each classification threshold. Each point on the ROC curve indicates the TPR and FPR at each corresponding classification threshold. Both TPR and FPR may increase as the classification threshold decreases. Line 1010 in Graph 1000 is the ROC of a random classifier. Point 1030 in the upper left corner represents an ideal classifier with a TPR of 1 and an FPR of 0. If the corresponding ROC curve is closer to the upper left corner, the classifier's performance is likely better. The performance of a classifier can be determined based on, for example, the area under the ROC curve (AUC). If the corresponding AUC of the classifier is close to 1, the classifier's performance is likely better.

[0100] existFigure 10 In the example shown, ROC curve 1020 is an example of the ROC curve for the classifier described above, such as classifying CGM data samples with known GMI results (>7% or ≤7%) into groups that meet or do not meet the target value of a TIR metric (e.g., TIR, TITR, TB70, TB54, TA180, or TA250). Region 1040 shows the classifier's AUC. To determine an appropriate classification threshold, point 1022, the closest point to point 1030 among all points on ROC curve 1020, can be determined, and the corresponding classification threshold for point 1022 can be determined as the corresponding target value of the TIR metric that corresponds to the target of the GMI metric (e.g., approximately 7%). Target values ​​corresponding to a certain target value of the CGM metric can be determined in the same manner based on the ROC curve of the corresponding classifier for other CGM metrics.

[0101] Figure 11 Table 1100 illustrates examples of target values ​​determined for certain time-range metrics within a range, corresponding to different GMI target values, according to certain implementation schemes. Time-range metrics may include, for example, TITR, TIR, TA180, TA250, TB70, TB54, etc. Different GMI target values ​​may include, for example, below 7%, below 6.8%, below 6.6%, below 6.5%, etc. Table 1100 also shows the AUC of the corresponding classifier and the corresponding sensitivity, specificity, and accuracy of the classifier at the determined target values ​​for the time-range metrics. Table 1100 shows results based on both CGM data samples from users aged 15 years or younger and CGM data samples from users older than 15 years.

[0102] In the illustrated example, for users aged 15 years or younger, when the GMI target is below 7%, the determined target values ​​for TITR (70 mg / dL to 140 mg / dL), TIR (70 mg / dL to 180 mg / dL), TA180 (>180 mg / dL), TA250 (>250 mg / dL), TB70 (<70 mg / dL), and TB54 (<54 mg / dL) are approximately greater than 46.9%, greater than 69.6%, less than 28.1%, less than 6.9%, greater than 2.4%, and greater than 0.4%, respectively. The determined target values ​​for TITR, TIR, TA180, and TA250 have relatively high sensitivity, specificity, and accuracy (e.g., about 90% or higher), but the determined target values ​​for TB70 and TB54 have relatively low sensitivity, specificity, and accuracy (e.g., between about 50% and about 70%). This can indicate a strong correlation between the GMI metric and the TITR, TIR, TA180, and TA250 metrics, and a weaker correlation between the GMI metric and the TB70 and TB54 metrics. When metrics (e.g., TB70 and TB54) have a weak correlation with the GMI metric, identifying reliable targets for the metric can be challenging.

[0103] For other GMI target values ​​(e.g., <6.8%, <6.6%, or <6.5%), the sensitivity, specificity, and accuracy of the target values ​​determined by TITR, TIR, TA180, and TA250 are also relatively high (e.g., approximately 90% or higher), but the sensitivity, specificity, and accuracy of the target values ​​determined by TB70 and TB54 are relatively low (e.g., between approximately 50% and approximately 70%). Specifically, for all GMI target values, the sensitivity, specificity, and accuracy of the target values ​​determined by TITR and TA180 are consistently greater than 90%, and for all GMI target values, the AUC of the classifier used to classify the TITR and TA180 measures based on the GMI target values ​​is close to 1 (e.g., ≥98%). For TIR and TA250, the sensitivity, specificity, and accuracy may be slightly lower. This may suggest that TITR or TA180 may be better measures in glucose management when the goal is to achieve normal blood glucose levels.

[0104] Similarly, for users older than 15 years, when the GMI target is below 7%, the target values ​​determined by TITR, TIR, TA180, TA250, TB70, and TB54 are approximately greater than 45.7%, greater than 71.8%, less than 26.7%, less than 5%, greater than 1.5%, and greater than 0.2%, respectively. The target values ​​determined by TITR, TIR, TA180, and TA250 have relatively high sensitivity, specificity, and accuracy (e.g., approximately 90% or higher), but the target values ​​determined by TB70 and TB54 have relatively low sensitivity, specificity, and accuracy (e.g., between approximately 50% and approximately 70%). For other GMI target values ​​(e.g., <6.8%, <6.6%, or <6.5%), the sensitivity, specificity, and accuracy of the target values ​​determined by TITR, TIR, TA180, and TA250 are also relatively high (e.g., approximately 90% or higher), but the sensitivity, specificity, and accuracy of the target values ​​determined by TB70 and TB54 are relatively low (e.g., between approximately 55% and approximately 70%). Specifically, for all GMI target values, the sensitivity, specificity, and accuracy of the target values ​​determined by TITR and TA180 are consistently greater than 90%, and for all GMI target values, the AUC of the classifiers used to classify the TITR and TA180 measures based on the GMI target values ​​is close to 1 (e.g., ≥98%). For TIR and TA250, the sensitivity, specificity, and accuracy may be slightly lower.

[0105] Figure 11 It is also shown that as GMI targets become more stringent (e.g., progressing from <7% to <6.5%), the reliability of the identified time-in-time targets associated with the GMI targets may decrease slightly, as evidenced by a decrease in sensitivity, specificity, accuracy values ​​(e.g., from about 90% to 92% to about 86% to 84%) and AUC values ​​(e.g., from about 97% to 98% to about 94%). Conversely, the reliability of the identified TITR targets associated with different GMI targets can always be high, regardless of the stringency of the GMI targets.

[0106] For example Figure 11 As illustrated, the identified targets for TB70 and TB54 may have the lowest values ​​for sensitivity, specificity, accuracy (e.g., as low as 50%) and AUC (e.g., as low as 61%), which may indicate that targets for time consumed within a low glycemic range may not be correlated with GMI targets very consistently. Further research into this discrepancy is described in more detail below.

[0107] The results in Table 1100 show that for a GMI target of less than 7%, the corresponding TIR target is greater than approximately 70% for users aged 15 or younger and greater than approximately 72% for users older than 15. For the same GMI target of less than 7%, the corresponding TA180 target is less than approximately 28% for users aged 15 or younger and less than approximately 27% for users older than 15, while the corresponding TA250 target is less than approximately 7% for users aged 15 or younger and less than approximately 5% for users older than 15. Even though the consensus limits for TA180 and TA250 are set at <25% and <5%, respectively, the results shown in Table 1100 indicate that a TA180 target between approximately 28% and 25% (for users aged 15 years or younger) or between approximately 26.7% and 25% (for users older than 15 years) corresponds to a GMI target of less than 7%, while a TA250 target between approximately 6.9% and 5% (for users aged 15 years or younger) or between approximately 5% and lower (for users older than 15 years) corresponds to a GMI target of less than 7%. Generally, these target values ​​are closely consistent with the recommended international consensus targets for time measures within the aforementioned ranges, indicating that the recommended consensus targets for GMI measures and time measures across various ranges of type 1 diabetes are generally consistent and accurately co-defined.

[0108] The results in Table 1100 further show that, for a GMI target of less than 7%, the associated TITR target is greater than approximately 47% for users aged 15 or younger, and greater than approximately 46% for users older than 15. For both age groups, the associated TITR target is greater than approximately 51% for a GMI target of less than 6.8%. Figure 8 As shown, a TITR target of more than 50% can be achieved by about 45% of users aged 15 or younger and about 46% of users older than 15, indicating that a TITR target of more than 50% is reasonable for both younger and older users.

[0109] The data used in the analysis also showed that Medtronic MiniMed TM The 780G system can achieve mean TB70 and mean TB54 values ​​that are significantly lower than clinical targets of <4% and <1%, respectively. Figure 8 and Figure 9 As shown. Figure 8 and Figure 9It was also shown that a lower mean GMI and a larger mean TIR, as well as a mean TITR, can be achieved by consistently using the recommended settings while maintaining TB70 and TB54 well below the recommended targets (e.g., mean TB70 of approximately 2.9% and 2.4% for younger and older users, respectively, and mean TB54 of approximately 0.6% and 0.4% for younger and older users, respectively). However, the results in Table 1100 suggest that a mean TB70 of less than 2% is unlikely for a GMI target below approximately 6.5.

[0110] The results in Table 1100 also quantify the relationships or trade-offs between meeting or exceeding targets for various CGM measures of glycemic control. For example, attempting to achieve a low mean GMI may be accompanied by a reduction in the duration of hyperglycemia and an increase in TIR and TITR. In one example, for users older than 15 years, when the GMI target is less than 6.6%, the corresponding TIR target can be greater than 80%, the corresponding TITR target can be greater than 56%, the corresponding TB70 target can be greater than 2%, and the corresponding TA180 target can be less than 17%.

[0111] Therefore, the techniques and results disclosed herein can be used to develop treatment strategies to improve TIR and TITR and reduce the duration of hypoglycemia. For example, to maintain a mean TB70 of less than 3%, a mean TIR greater than 70% is achievable for users aged 15 years or younger, but may not exceed 78% (or a TITR greater than 55%). A TIR greater than 78% is possible, but may be accompanied by a TB70 greater than 3% (still below the recommended target of <4%). For users older than 15 years, an 81% TIR (or a TITR of 58%) is achievable at the cost of a TB70 greater than 2.3%. Typically, a trade-off is seen in both age groups, but older age groups may be more likely to achieve stricter glycemic targets. These techniques and results enable healthcare providers and individuals with type 1 diabetes to develop realistic expectations regarding overall glycemic control performance when pursuing stricter time targets across various ranges and GMI measures using the AID system.

[0112] The examples described above illustrate the target of a population-based CGM metric (e.g., a time-in-range metric or a GMI metric) and its relationship with a specific glycemic control system (such as Medtronic MiniMed). TMThis technique describes the methods and results of determining a target for another CGM metric (e.g., another time-in-range metric or GMI metric) based on a population CGM data sample associated with a 780G (780G) control system. Similar techniques can be used to determine a target for another CGM metric based on a target for one CGM metric in a population and / or on a population CGM data sample associated with another blood glucose control system. Similar techniques can also be used to determine a target for another CGM metric for a specific user based on a target for one CGM metric and an individual CGM data sample associated with a specific blood glucose control system. For example, CGM data for a user of a specific blood glucose control system can be collected over a period of time and divided into multiple portions corresponding to multiple CGM data samples (e.g., each portion includes data for 10 to 14 days or longer), and the GMI and TIR metrics for each CGM data sample can be determined and used as training or testing samples to determine a target for another CGM metric based on a user's target for one CGM metric using the techniques described above.

[0113] As mentioned above, for example Figure 11 The results shown indicate that stronger associations may exist between some CGM measures and weaker associations between others. To determine meaningful objectives for time measures within the range in relation to GMI objectives, it may be desirable to determine the associations between GMI and time measures within the range (such as TIR, TITR, TB70, TB54, TA180, and TA250). According to certain implementations, dimensionality or parameter reduction techniques (e.g., principal component analysis (PCA)) and correlation techniques (e.g., pairwise Pearson correlation) can be used to determine the associations between GMI and various time measures within the range based on the aforementioned CGM data sample, and to reduce the dimensionality of these CGM measures while preserving relevant information inherent in the CGM data sample (e.g., most of the variance). For example, time measures within the range and GMI measures can be clustered to identify measures associated with GMI, such as those belonging to the same dimension or orthogonal to GMI, thereby validating the determination of objectives within the various time ranges associated with GMI objectives.

[0114] The Pearson correlation coefficient measures the linear correlation between two variables and has values ​​between -1 and 1. This value measures the strength and direction of the relationship between the two variables. A Pearson correlation coefficient close to zero indicates no relationship or a weak relationship between the two variables. A Pearson correlation coefficient between 0 and 1 indicates a positive correlation, where a change in one variable may lead to a change in the other variable in the same direction. A Pearson correlation coefficient between 0 and -1 indicates a negative correlation, where a change in one variable may lead to a change in the other variable in the opposite direction.

[0115] Figure 12A andFigure 12B Examples of Pearson correlation analysis results for time measures and GMI measures across six ranges are shown for users aged 15 or younger and users aged 15 and older, respectively. Figure 12A and Figure 12B In the diagram, darker boxes indicate stronger positive or negative pairwise correlations. The value in each box indicates the corresponding Pearson correlation coefficient. Larger positive Pearson correlation coefficients (e.g., closer to 1, such as 0.98 between GMI and TA180) indicate a stronger positive correlation, while smaller negative Pearson correlation coefficients (e.g., closer to -1, such as -0.98 or -0.99 between TIR and TA180) indicate a stronger negative correlation. Pearson correlation coefficients closer to zero (e.g., approximately 0 or -0.05 between TA250 and TB54, or approximately -0.07 or 0.02 between TIR and TB54) indicate a weak or no correlation. Figure 12A and Figure 12B The results show CGM measures for two different groups across two age groups, where measures within each group can have high within-group correlation and low between-group correlation for both age groups. For example, measures of GMI, TITR, TIR, TA180, and TA250 can be strongly correlated with each other, but weakly correlated with TB70 and TB54 measures, which can be strongly correlated. In the first group, TITR, TIR, TA180, and TA250 all showed strong correlations with GMI (e.g., absolute correlation > 0.89). The second group includes TB70 and TB54 measures associated with hypoglycemia. Each of TB70 and TB54 measures can have a low correlation with GMI (e.g., absolute correlation < 0.43), indicating that measures of time spent in the hypoglycemic range may be misaligned on the same dimension as GMI measures.

[0116] PCA is a statistical technique in the field of multivariate data analysis designed to explain the maximum variance using a minimal number of components. In this paper, PCA can be used to reduce the dimensionality and complexity of these CGM measures while preserving relevant information in the dataset (e.g., most of the variance). Using PCA, time-in-range measures aligned with GMI measures within the same principal components and therefore part of the same dimension can be identified. Identifying time-in-range measures aligned with GMI measures highlights those closely related to GMI measures and justifies the determination of the time-in-range objective based on the aforementioned GMI objective. For time-in-range measures in dimensions different from (e.g., orthogonal to) GMI measures, it may be necessary to set objectives based on the objectives of other relevant CGM measures in the same dimension as such time-in-range measures.

[0117] Figure 12C and Figure 12D The results of PCA analyses of time-based and GMI measures are presented separately for users aged 15 or younger and users aged 15 and older, across six ranges. Figure 12C and Figure 12D In the example shown, the Kaiser criterion guideline is used to reduce the dimensionality of these CGM measures to two, where the horizontal axis represents the first principal component explaining the largest variance, and the vertical axis represents the second principal component explaining the second largest variance. Each principal component can be a new transformed variable, which is a linear combination of the original variables. Figure 12C and Figure 12D Each vector in the vector indicates the contribution (or load) of the blood glucose metric to the two principal components or dimensions. Figure 12C and Figure 12D The results show that for both age groups, the TITR, TIR, TA180, and TA250 measures closely correlated with the GMI measure within the first principal component. The second principal component consisted primarily of the hypoglycemic measures TB70 and TB54.

[0118] Figure 12C The diagram shows that two dimensions, or principal components, explain the majority of the total variance for users aged 15 or younger (e.g., approximately 95.7%), with the first principal component explaining approximately 68.6% of the variance and the second principal component explaining approximately 27.1% of the variance. Figure 12D The results show that two principal components explain approximately 95.2% of the total variance for users older than 15 years, with the first principal component explaining approximately 69.1% and the second principal component explaining approximately 26.1%.

[0119] The Pearson correlation and PCA analyses described above consistently placed TITR, TIT, TA180, and TA250 in the same dimension as GMI, and TB70 and TR54 in a different dimension than GMI. (As described above...) Figure 11 As shown in Table 1100, the lowest sensitivity, specificity, accuracy, and AUC were correlated with the TB70 and TB54 targets for any given GMI target, indicating a weak association between time and GMI measures in the hypoglycemic range. Results from Pearson correlation and PCA analyses confirmed that targets for TIR, TITR, TA180, and TA250 can be reliably determined based on GMI targets.

[0120] The results of the Pearson correlation and PCA analyses described above further confirm that the TB70 and TB54 measures measure dimensions of glycemic control associated with hypoglycemia, and that the GMI measure is not aligned in this dimension. This indicates that targets for these hypoglycemic measures may not be reliably determined based solely on GMI targets. Therefore, stricter glycemic control (e.g., lower GMI or HbA1c) may not necessarily imply an increase in time below range (TBR), and reducing TBR may not necessarily lead to a lower time relative intensity (TIR) ​​or a higher GMI. For example, reducing time above range (TAR) can improve GMI and can achieve a reduction in TBR without worsening (increasing) GMI. Since the TB70 and TB54 measures are aligned in the same dimension, targets for one of TB54 and TB70 can be reliably estimated based on targets for the other.

[0121] Figure 13 Table 1300 is included, which shows examples of estimated target results for TB54 associated with various TB70 targets according to certain implementation schemes. Figure 13 The results in can be used to determine Figure 11 The results shown were determined using techniques similar to those used in the study. In Table 1300, the targets for the TB70 metric used to estimate the TB54 target include below 4%, below 3.5%, below 3%, and below 2.5%. Figure 13 The results show that for both age groups, a TB70 target of less than 4% corresponds to a TB54 target of less than 0.73%, while for users aged 15 years or younger and users older than 15 years, a TB70 target of less than 2.5% corresponds to a TB54 target of less than 0.45% and 0.43%, respectively. The AUC is relatively high for both age groups, and the sensitivity, specificity, and accuracy for both age groups and all four TB70 targets are better than 85%.

[0122] The techniques and results described above can be used to establish and validate the appropriateness of consensus targets for glycemic control. The results reveal and quantify the trade-offs between targets for various CGM measures, allowing users to know the actual expectations for other CGM measures when setting targets for CGM measures on an AID system. For example, corresponding targets for CGM measures can be determined for the AID system and groups or specific users as described above, and users can be provided with a user application to convert the target of a first CGM measure entered or selected by the user into estimated corresponding targets for other CGM measures, allowing users to know the expected performance of the AID system in other CGM measures. If the estimated corresponding target of another CGM measure does not meet the expected glucose control performance, the user can select a different target for the first CGM measure and determine whether the estimated corresponding targets of other CGM measures meet the expected glucose control performance. In this way, different CGM glycemic measures can be considered for glycemic control using an AID system. Furthermore, the techniques disclosed herein can be used, for example, to support and form a basis for recommending consensus targets for new CGM measures (e.g., TITR or another consensus target for a CGM measure) based on generally agreed-upon targets for other CGM measures (e.g., GMI or TIR).

[0123] The techniques disclosed herein can be used to determine associations between different CGM measures for a group or a specific user. These techniques can be used to determine associations between different CGM measures across different AID systems and can also be applied to applications beyond type 1 diabetes and AID systems. For example, the techniques disclosed herein can also be used to determine associations and dimensions of CGM measures using various datasets, including data on individuals with type 1 or type 2 diabetes, individuals receiving multiple daily injections, individuals using only CGM, individuals using sensor-enhanced pump therapy, etc.

[0124] Figure 14 Flowchart 1400 includes examples of methods for determining a target value corresponding to a target value of another CGM target, according to certain implementations. The operations in flowchart 1400 can be performed using, for example, one or more processors or computing systems (e.g., a server, computing system 1600 described below, or a smart user device). Although flowchart 1400 may describe the operations as a sequential sequence, some operations may be performed in parallel or concurrently. Additionally, the order of operations may be rearranged. Processes may have additional steps not included in the flowchart. In some specific implementations, some operations may be optional or may be omitted. Some operations may be performed more than once.

[0125] exist Figure 14In the illustrated example, the operations in flowchart 1400 may include: at box 1410, using a classifier to classify CGM data samples that meet and do not meet the target value of a first CGM metric into samples that meet or do not meet a plurality of threshold targets of a second CGM metric. As described above, the classifier may be a binary classifier and may include, for example, a logistic regression model, a support vector machine, a neural network, a decision tree, etc. CGM data samples may include, for example, CGM data samples of users or CGM data samples of a population. CGM data samples may include CGM data samples of users in an age group. In one example, CGM data samples may include CGM data samples of users of an automated insulin delivery system. In one example, CGM data samples may include CGM data samples of users with type 1 diabetes and / or users with type 2 diabetes. In one example, each CGM data sample may include CGM data of a user over a period of time (e.g., 10 to 14 days or longer). In some examples, a CGM metric may be determined for each CGM data sample and these metrics may be associated with the CGM data sample. In some examples, CGM data samples may be divided into training samples and test samples. The first CGM metric may include, for example, GMI, TIR, TITR, TA180, TA250, TB70, or TB54. The second CGM metric may include, for example, GMI, TIR, TITR, TA180, TA250, TB70, or TB54. In one example, the target value for the first CGM metric may be a GMI below 7%, 6.8%, 6.6%, or 6.5%, and the second CGM metric may include TIR, TITR, TA180, TA250, TB70, or TB54.

[0126] The operation in box 1420 may include: for each of the plurality of threshold targets of the second CGM metric, determining the true positive rate (TPR) and false positive rate (FPR) of the classifier based on the classification results. The TPR may be the ratio between the number of true positive categories and the total number of true positive and false negative categories (e.g., TPR = TP / (TP + FN)). The FPR may be the ratio between the number of false positive categories and the total number of false positive and true negative categories (e.g., FPR = FP / (FP + TN)).

[0127] The operations in boxes 1430 to 1450 may include: identifying at least the target value of the second CGM metric based on the TPR and FPR of the classifier for each of the plurality of threshold targets of the second CGM metric. For example, at box 1430, a receiver operating characteristic (ROC) curve of the classifier may be generated for each threshold target based on the TPR and FPR of the classifier. Each point of the ROC curve may correspond to a different threshold target among the plurality of threshold targets of the second CGM metric and may have a corresponding TPR and FPR.

[0128] The operation in box 1440 may include: selecting a first data point of the ROC curve that is closest to a point where the TPR is one and the FPR is zero. Selection may be based on the distance between the top-left corner of the ROC curve and a point on the ROC curve. The selected data point may have both high TPR and low FPR.

[0129] The operation in box 1450 may include: identifying the threshold target associated with the first data point as the target value of the second CGM metric corresponding to the first target value of the first CGM metric.

[0130] Figure 15 This includes flowchart 1500, which illustrates an example of a processor implementation method for providing a user with a target value corresponding to a target value of another CGM target, according to certain embodiments. The operations in flowchart 1500 can be performed using, for example, one or more processors or computing systems (e.g., a server, computing system 1600 described below, or intelligent user devices such as smartphones). Although flowchart 1500 may describe the operations as a sequential sequence, some operations may be performed in parallel or concurrently. Additionally, the order of operations may be rearranged. The process may have additional steps not included in the flowchart. In some specific embodiments, some operations may be optional or may be omitted. Some operations may be performed more than once. In some examples, the processor-implemented method may be implemented as a user application executable on an intelligent user device such as a smartphone.

[0131] exist Figure 15 In the illustrated example, the operation in flowchart 1500 may include, at block 1510, receiving input of a target value for a first continuous glucose monitoring (CGM) metric. As described above, the first CGM metric may include GMI, TIR, TITR, TA180, TA250, TB70, or TB54. For example, the first CGM metric may be GMI, and the target value for the first CGM metric may be below 7%, below 6.8%, below 6.6%, below 6.5%, below 6.4%, below 6.2%, below 6%, or even lower.

[0132] The operation in box 1520 may include: estimating a target value for at least a second CGM metric corresponding to the target value of the first CGM metric. The method for estimating the target value of the at least second CGM metric corresponding to the target value of the first CGM metric may be similar to that described above. Figure 14 The method described.

[0133] The operation in box 1530 may include providing the user with an estimated target value for at least the second CGM metric. In some examples, the method may further include: at box 1540, determining, for example, based on user feedback, whether the estimated target value for at least the second CGM metric meets a predetermined criterion; and at box 1550, configuring the insulin delivery system based on the target value of the first CGM metric if the estimated target value for at least the second CGM metric meets the predetermined criterion. If the estimated target value for at least the second CGM metric does not meet the predetermined criterion, the user may be prompted to enter or select an alternative target value.

[0134] Figure 16 This is a block diagram of an example computing system 1600 that can implement some of the examples disclosed herein. For example, computing system 1600 can be used to implement delivery device 102, monitoring device 104, computing device 106 and remote / cloud computing system 108, glucose sensor subsystem 210, controller 220, insulin delivery subsystem 230, glucose delivery subsystem 240 and glucagon delivery subsystem 250, CGM system 300, glucose sensor device 400, delivery device 500, or another device that can be used to calculate insulin dosage and / or predict future blood glucose levels. It should be noted that... Figure 16 This is intended only to provide a generalized illustration of the various components; any or all of these components may be used appropriately. Additionally, it may be noted that... Figure 16 The illustrated components can be located in a single device and / or predict future blood glucose levels. These are distributed across various networked devices, which can be located in different geographical locations.

[0135] In the illustrated example, computing system 1600 may include one or more processors 1610 and memory 1620. Processor 1610 may be configured to execute instructions stored in memory 1620 to perform one or more of the methods described herein and in other applications. Processor 1610 may include, for example, one or more central processing units, microprocessors, microcontrollers, special-purpose processors (e.g., digital signal processors), ASICs, DSPs, FPGAs, or other processors suitable for implementation within portable electronic devices. Processor 1610 may be communicatively coupled to multiple components within computing system 1600. To achieve this communicative coupling, processor 1610 may communicate with other illustrated components across bus 1640. Bus 1640 may be any subsystem suitable for transmitting data within computing system 1600. Bus 1640 may include multiple computer buses and additional circuitry for transmitting data.

[0136] Memory 1620 may include one or more transient and / or non-transitory storage devices, such as, for example, static random access memory (SRAM), dynamic random access memory (DRAM), read access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), FLASH-EPROM, secure digital card (SD) cards, and any other memory chip or enclosure. Such storage devices may be configured to implement any suitable data storage, including but not limited to various file systems, data structures, computer-readable instructions, program modules, etc. In some embodiments, memory 1620 may be distributed across different hardware modules. Instruction sets and / or code sets may be stored on memory 1620. These instructions may take the form of executable code executable by computing system 1600, and / or may take the form of source code and / or installable code, which take the form of executable code when compiled and / or installed on computing system 1600 (e.g., using any of a variety of generally available compilers, installers, compression / decompression utilities, etc.).

[0137] Memory 1620 may include an operating system 1625 loaded therein. Operating system 1625 is operable to initiate the execution of instructions provided by application modules 1622 through 1624 and / or manage other hardware modules 1670, and to interface with a communications subsystem 1630, which may include one or more wired and / or wireless transceivers. Operating system 1625 may be adapted to perform other operations across components of computing system 1600, including threading, virtualization, resource management, data storage control, and other similar functionalities. In some embodiments, memory 1620 may store multiple application modules 1622 through 1624, which may include any number of applications. Examples of applications may include an insulin calculator, a blood glucose level monitor or predictor, a glucose level management application, etc. Application modules 1622 through 1624 may include specific instructions to be executed by processor 1610. In some embodiments, certain applications or portions of application modules 1622 through 1624 may be executed by other hardware modules 1670.

[0138] The communication subsystem 1630 may include, for example, infrared communication devices, wireless communication devices, and / or chipsets (such as...) The communication subsystem 1630 includes devices such as IEEE 802.11 devices, Wi-Fi devices, WiMax devices, cellular communication devices, etc., and / or similar communication interfaces. One or more antennas (not shown) may be used for wireless communication as part of the communication subsystem 1630 or as separate components (such as wireless charging receivers or near-field communication receivers) coupled to any part of the computing system 1600. In some embodiments, the communication subsystem 1630 may include circuitry for wired communication technologies such as Ethernet, coaxial communication, Universal Serial Bus (USB), etc. In some embodiments, the communication subsystem 1630 may include transceivers for communicating with base transceiver stations and other wireless devices and access points, which may include communication with different data networks and / or network types, such as wide area networks (“WAN”), wireless wide area networks (WWAN), local area networks (LAN), wireless local area networks (WLAN), personal area networks (PAN), or wireless personal area networks (WPAN). A WWAN may be, for example, a WiMax (IEEE 802.16) network. A WLAN may be, for example, an IEEE 802.11x network. WPAN can be, for example, a Bluetooth network, IEEE 802.15x, or some other type of network. The techniques described herein can also be used in any combination of WAN, LAN, PAN, WWAN, WLAN, and / or WPAN. The communication subsystem 1630 can permit the exchange of data with networks, other computer systems, and / or any other devices. For example, the communication subsystem 1630 can be used to receive a treatment determination, such as receiving a therapeutic fluid (e.g., insulin) delivery from a cloud computing system via an intermediate computing device (e.g., a controller) communicatively coupled to the computing system 1600, whereby the processor 1610 can, based on the treatment determination, transmit commands to an actuator controller to induce the delivery of an appropriate amount of therapeutic fluid (e.g., insulin) to a user. In another example, the communication subsystem 1630 can be used to transmit measurement results (e.g., sensor glucose levels) to a computing device (e.g., a smartphone or personal health monitoring device) and / or via the computing device to a remote server, or to receive data (e.g., calibration data, configuration data, etc.) from or via a computing device or remote server.

[0139] The computing system 1600 may include a display module 1650. The display module 1650 can present information from the computing system 1600 to a user, such as text, images, audio, video, and various commands. This information may be derived from one or more application modules 1622 to 1624, a communication / networking subsystem 1630, one or more other hardware modules 1670, combinations thereof, or any other suitable means. For example, the display module 1650 may be used to display user queries, which may include text, images, waveforms, audio clips, video clips, etc. The display module 1650 may use liquid crystal display (LCD) technology, light-emitting diode (LED) technology (including, for example, OLED, ILED, LED, AMOLED, TOLED, etc.), light-emitting polymer display (LPD) technology, or some other display technology.

[0140] The input / output user interface 1660 allows a user to send action requests to the computing system 1600 to perform specific actions and can provide the user with information (e.g., the status of the computing system 1600, measurement results, alarms, etc.). The input / output user interface 1660 may include one or more input devices, such as a touchscreen, touchpad, microphone, button, dial pad, switch, keyboard, mouse, game controller, or any other suitable device for receiving action requests and transmitting the received action requests to the processor 1610. In some embodiments, the input / output user interface 1660 may include one or more output devices, such as a display, speaker, light-emitting device, haptic device, etc., to provide feedback or alarms to the user.

[0141] In some embodiments, the computing system 1600 may include a plurality of other hardware modules 1670. Each of the other hardware modules 1670 may be a physical module within the computing system 1600. While each of the other hardware modules 1670 may be permanently configured as a structure, some of the other hardware modules 1670 may be temporarily configured to perform a specific function or be temporarily activated. Examples of other hardware modules 1670 may include, for example, audio output and / or input modules (e.g., microphones or speakers), near field communication (NFC) modules, rechargeable batteries, battery management systems, wired / wireless battery charging systems, actuator controllers, etc. In some embodiments, one or more functions of the other hardware modules 1670 may be implemented in software.

[0142] In one example, the computing system 1600 may be part of an insulin delivery device (e.g., a pump) that delivers rapid-acting insulin via a tubing configured for fluid connection to a subcutaneously inserted cannula. The computing system 1600 may cause the delivery of two types of doses—a basal dose, which may be delivered periodically (e.g., every five minutes) throughout the day and night in micro-dose doses; and a bolus dose that covers and / or corrects for hyperglycemia caused by meals. The insulin delivery device may include a user interface with button elements that can be manipulated to administer insulin boluses, change therapy settings, change user preferences, and select display features, etc. The insulin delivery device may also include a display device that can be used to present various types of information or data to the user. According to various aspects of this disclosure, a user of the insulin delivery device may use the button elements to input certain event data (e.g., event type, event start time, event details, etc.) and may use the display device to confirm the user input.

[0143] In various specific implementations, the aforementioned hardware and modules may be implemented on a single device or on multiple devices that can communicate with each other using wired or wireless connections. In alternative configurations, different and / or additional components may be included in the computing system 1600. Similarly, the functionality of one or more components may be distributed among the components in a manner different from that described above.

[0144] Implementations of the methods disclosed herein can be implemented by hardware, software, firmware, middleware, microcode, hardware description languages, or any combination thereof. For example, the operation can be performed by one or more servers or other computer systems. When implemented in software, firmware, middleware, or microcode, program code or code segments for performing associated tasks can be stored in one or more computer-readable media, such as storage media, and can be executed by one or more processors to perform the associated tasks. Computer-readable media can include transient or non-transitory computer-readable media, such as RAM, ROM, EEPROM, flash memory, solid-state drives, hard disk drives, CDs, DVDs, or any other media that can be used to store desired program code in the form of instructions or data structures and is accessible by a computer. One or more processors can include general-purpose microprocessors, application-specific integrated circuits (ASICs), graphics processing units (GPUs), network processing units (NPUs), digital signal processors (DSPs), field-programmable arrays (FPGAs), etc.

[0145] The methods, systems, and apparatus discussed above are examples. Various procedures or components may be appropriately omitted, substituted, or added in different embodiments. For example, in alternative configurations, the described methods may be performed in a different order than described, and / or stages may be added, omitted, and / or combined. Furthermore, features described with respect to certain embodiments may be combined in various other embodiments. Different aspects and elements of embodiments may be combined in similar ways. Moreover, technology is constantly evolving, therefore many elements are examples and the scope of this disclosure is not limited to those specific examples.

[0146] Specific details are provided in the description to offer a thorough understanding of the embodiments. However, embodiments may be practiced without these specific details. For example, well-known circuits, processes, systems, structures, and techniques have been shown without the need for unnecessary detail to avoid obscuring the embodiments. This description provides only exemplary embodiments and is not intended to limit the scope, applicability, or configuration of the invention. Rather, the foregoing description of the embodiments will provide those skilled in the art with an enabling description for implementing various embodiments. Various changes may be made to the function and arrangement of the elements without departing from the spirit and scope of this disclosure.

[0147] Furthermore, some implementations are described as processes depicted as flowcharts or block diagrams. Although each process can be described as a sequential sequence of operations, many operations can be executed in parallel or concurrently. Additionally, the order of operations can be rearranged. Processes may have additional steps not included in the diagram. Furthermore, implementations of these methods can be implemented by hardware, software, firmware, middleware, microcode, hardware description languages, or any combination thereof. When implemented in software, firmware, middleware, or microcode, program code or code segments used to perform the associated tasks can be stored in a computer-readable medium such as a storage medium. The processor can execute the associated tasks.

[0148] It will be apparent to those skilled in the art that substantial variations can be made to suit specific requirements. For example, custom or dedicated hardware may be used, and / or specific elements may be implemented in hardware, software (including portable software such as applets), or both. Furthermore, connectivity with other computing devices, such as network input / output devices, may be employed.

[0149] Any of the techniques, operations, methods, programs, algorithms, or code described herein can be converted into or expressed in a programming language or computer program embodied on a computer, processor, or machine-readable medium. As used herein, the terms "programming language" and "computer program" each include any language used to specify instructions for a computer or processor, and include (but are not limited to) the following languages ​​and their derivatives: assembler, Basic, batch files, BCPL, C, C+, C++, Delphi, Fortran, Java, JavaScript, machine code, operating system command languages, Pascal, Perl, PL1, Python, scripting languages, visual Basic, meta-languages ​​that specify their own programs, and all first-, second-, third-, fourth-, fifth-, or higher generation computer languages. Databases and other data schemas, as well as any other meta-languages, are also included. No distinction is made between interpreted, compiled languages, or languages ​​that use both compilation and interpretation methods. No distinction is made between compiled and source versions of a program. Therefore, a reference to a program that can exist in more than one state (such as source, compiled, target, or linked) is a reference to any and all such states. A reference to a program can encompass the actual instructions and / or the intent of those instructions.

[0150] Referring to the accompanying drawings, components that may include memory may include non-transitory machine-readable media. The terms "machine-readable media" and "computer-readable media" can refer to any storage medium involved in providing data that enables a machine to operate in a particular manner. In the embodiments provided above, various machine-readable media may involve providing instructions / code to a processing unit and / or other devices for execution. Additionally or alternatively, machine-readable media may be used to store and / or carry such instructions / code. In many embodiments, computer-readable media are physical and / or tangible storage media. Such media can take many forms, including but not limited to non-volatile media, volatile media, and transmission media. Common forms of computer-readable media include, for example, magnetic and / or optical media such as optical discs (CDs) or digital multifunction discs (DVDs), punched cards, paper tape, any other physical media with a perforated pattern, RAM, programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), FLASH-EPROM, any other memory chip or cartridge, a carrier wave as described below, or any other medium from which a computer can read instructions and / or code. Computer program products may include code and / or machine-executable instructions, which may represent any combination of procedures, functions, subroutines, programs, routines, application programs (Apps), subroutines, modules, software packages, classes, or instructions, data structures, or program statements.

[0151] Those skilled in the art will understand that information and signals used to transmit the messages described herein can be represented using any of a variety of different technologies and processes. For example, data, instructions, commands, information, signals, bits, symbols, and chips referenced throughout the foregoing description can be represented by voltage, current, electromagnetic waves, magnetic fields or magnetic particles, optical fields or optical particles, or any combination thereof.

[0152] As used herein, the terms “and” and “or” can have a variety of meanings, which are also expected to depend at least in part on the context in which the terms are used. Generally, when used with a list of related terms (such as A, B, or C), “or” is intended to mean A, B, and C (used herein in an inclusive sense) and A, B, or C (used herein in an exclusive sense). Furthermore, the term “one or more” as used herein can be used to describe any feature, structure, or property of the singular, or to describe some combination of features, structures, or properties. However, it should be noted that this is merely an illustrative example, and the claimed subject matter is not limited to this example. Additionally, when used with a list of related terms (such as A, B, or C), the term “at least one” can be interpreted as meaning A, B, C, or any combination of A, B, and / or C, such as AB, AC, BC, AA, ABC, AAB, AABBCCC, etc.

[0153] Furthermore, while certain embodiments have been described using specific combinations of hardware and software, it should be recognized that other combinations of hardware and software are also possible. Some embodiments may be implemented using only hardware, or only software, or a combination thereof. In one example, the software may be implemented using a computer program product comprising computer program code or instructions executable by one or more processors to perform any or all of the steps, operations, or processes described in this disclosure, wherein the computer program may be stored on a non-transitory computer-readable medium. The various processes described herein may be implemented on the same processor or different processors in any combination.

[0154] When a device, system, component, or module is described as being configured to perform certain operations or functions, such configuration can be achieved, for example, by designing electronic circuits to perform the operations, by programming programmable electronic circuits (such as microprocessors) to perform the operations (such as by executing computer instructions or code), or by a processor or core programmed to execute code or instructions stored on a non-transitory memory medium, or any combination thereof. Processes can communicate using a variety of technologies, including but not limited to conventional technologies for inter-process communication, and different pairs of processes can use different technologies, or the same pair of processes can use different technologies at different times.

[0155] Therefore, the specification and drawings should be viewed in an illustrative rather than restrictive sense. However, it will be apparent that additions, deletions, omissions, and other modifications and changes may be made therein without departing from the broader spirit and scope set forth in the claims. Thus, although specific embodiments have been described, these embodiments are not intended to be limiting. Various modifications and equivalents are within the scope of the appended claims.

[0156] Based on this description, implementations may include different combinations of features. Examples of implementations are described in the following numbered clauses:

[0157] Clause 1. A processor-implemented method, the processor-implemented method comprising:

[0158] Receive the target value as input for the first continuous glucose monitoring (CGM) metric;

[0159] Estimate the target value of at least a second CGM metric corresponding to the target value of the first CGM metric; and

[0160] Provide the user with the estimated target value of at least the second CGM metric.

[0161] Clause 2. The processor-implemented method according to Clause 1, wherein the first CGM metric comprises:

[0162] Glucose Management Index (GMI);

[0163] The percentage of time (TIR) ​​during which glucose levels range from 70 mg / dL to 180 mg / dL;

[0164] The percentage of time during which glucose levels were in the range of 70 mg / dL to 140 mg / dL (TITR);

[0165] The percentage of time during which glucose levels were above 180 mg / dL (TA180);

[0166] The percentage of time during which glucose levels were above 250 mg / dL (TA250);

[0167] The percentage of time during which glucose levels were below 70 mg / dL (TB70); or

[0168] The percentage of time during which glucose levels were below 54 mg / dL (TB54).

[0169] Clause 3. The processor-implemented method according to Clause 1 or 2, wherein the second CGM metric includes GMI, TIR, TITR, TA180, TA250, TB70, or TB54.

[0170] Clause 4. A processor-implemented method according to any one of Clauses 1 to 3, wherein estimating the target value of at least the second CGM metric corresponding to the target value of the first CGM metric comprises:

[0171] The classifier is used to classify the CGM data samples that satisfy and do not satisfy the target value of the first CGM metric into samples that satisfy or do not satisfy the second CGM metric for each of a plurality of threshold targets.

[0172] For each of the plurality of threshold targets quantified by the second CGM, the true positive rate (TPR) and false positive rate (FPR) of the classifier are determined; and

[0173] For each of the plurality of threshold targets in the second CGM metric, at least the target value of the second CGM metric is identified based on the TPR and FPR of the classifier.

[0174] Clause 5. The processor-implemented method according to Clause 4, wherein identifying at least the target value of the second CGM metric based on the TPR and FPR of the classifier for each of the plurality of threshold targets of the second CGM metric comprises:

[0175] For each of the plurality of threshold targets measured by the second CGM, the receiver operating characteristic (ROC) curve of the classifier is determined based on the TPR and FPR of the classifier;

[0176] Select the first data point of the ROC curve, the first data point being closest to the point where TPR is 1 and FPR is zero; and

[0177] The threshold target associated with the first data point is identified as the target value of the second CGM metric corresponding to the target value of the first CGM metric.

[0178] Clause 6. The method implemented by the processor according to Clause 4 or 5, wherein the CGM data sample includes at least one of the following:

[0179] The user's CGM data sample;

[0180] CGM data sample of the group;

[0181] CGM data samples from users in age groups;

[0182] CGM data sample from users of automated insulin delivery systems;

[0183] CGM data samples from users with type 1 diabetes and / or users with type 2 diabetes; or

[0184] Training samples and test samples.

[0185] Clause 7. A method implemented by a processor according to any one of Clauses 4 to 6, wherein the classifier includes a binary classifier.

[0186] Clause 8. The method implemented by the processor according to any one of Clauses 1 to 7, wherein the method implemented by the processor further comprises:

[0187] Determine that the estimated target value of at least the second CGM metric satisfies a predetermined criterion; and

[0188] Configure the insulin delivery system based on the target value of the first CGM metric.

[0189] Clause 9. A processor-implemented method, the processor-implemented method comprising:

[0190] The classifier is used to classify continuous glucose monitoring (CGM) data samples that meet and do not meet the target value of the first CGM metric into samples that meet or do not meet the target of the second CGM metric.

[0191] For each of the plurality of threshold targets quantified by the second CGM, the true positive rate (TPR) and false positive rate (FPR) of the classifier are determined; and

[0192] For each of the plurality of threshold targets in the second CGM metric, at least the target value of the second CGM metric is identified based on the TPR and FPR of the classifier.

[0193] Clause 10. The processor-implemented method according to Clause 9, wherein identifying at least the target value of the second CGM metric based on the TPR and FPR of the classifier for each of the plurality of threshold targets of the second CGM metric comprises:

[0194] For each of the plurality of threshold targets measured by the second CGM, the receiver operating characteristic (ROC) curve of the classifier is determined based on the TPR and FPR of the classifier;

[0195] Select the first data point of the ROC curve, the first data point being closest to the point where TPR is 1 and FPR is zero; and

[0196] The threshold target associated with the first data point is identified as the target value of the second CGM metric corresponding to the target value of the first CGM metric.

[0197] Clause 11. The processor-implemented method according to Clause 9 or 10, wherein the first CGM metric comprises:

[0198] Glucose Management Index (GMI);

[0199] The percentage of time (TIR) ​​during which glucose levels range from 70 mg / dL to 180 mg / dL;

[0200] The percentage of time during which glucose levels were in the range of 70 mg / dL to 140 mg / dL (TITR);

[0201] The percentage of time during which glucose levels were above 180 mg / dL (TA180);

[0202] The percentage of time during which glucose levels were above 250 mg / dL (TA250);

[0203] The percentage of time during which glucose levels were below 70 mg / dL (TB70); or

[0204] The percentage of time during which glucose levels were below 54 mg / dL (TB54).

[0205] Clause 12. The method implemented by the processor according to any one of Clauses 9 to 11, wherein the second CGM metric includes GMI, TIR, TITR, TA180, TA250, TB70, or TB54.

[0206] Clause 13. The method implemented by the processor according to any one of Clauses 9 to 12, wherein the CGM data sample comprises at least one of the following:

[0207] User's CGM data sample;

[0208] CGM data sample of the group;

[0209] CGM data samples from users in age groups;

[0210] CGM data samples from users of automated insulin delivery systems; or

[0211] CGM data sample of users with type 1 diabetes and / or users with type 2 diabetes.

[0212] Clause 14. The method implemented by the processor according to any one of Clauses 9 to 13, wherein the CGM data sample includes CGM data samples of users of an automated insulin delivery system.

[0213] Clause 15. A system comprising:

[0214] One or more processors;

[0215] One or more processor-readable storage media storing instructions that, when executed by the one or more processors, cause operations including:

[0216] Receive the target value as input for the first continuous glucose monitoring (CGM) metric;

[0217] Estimate the target value of at least a second CGM metric corresponding to the target value of the first CGM metric; and

[0218] Provide the user with the estimated target value of at least the second CGM metric.

[0219] Clause 16. The system according to Clause 15, wherein estimating the target value of at least the second CGM metric corresponding to the target value of the first CGM metric comprises:

[0220] The classifier is used to classify the CGM data samples that satisfy and do not satisfy the target value of the first CGM metric into samples that satisfy or do not satisfy the second CGM metric for each of a plurality of threshold targets.

[0221] For each of the plurality of threshold targets quantified by the second CGM, the true positive rate (TPR) and false positive rate (FPR) of the classifier are determined; and

[0222] For each of the plurality of threshold targets in the second CGM metric, at least the target value of the second CGM metric is identified based on the TPR and FPR of the classifier.

[0223] Clause 17. The system according to Clause 16, wherein identifying at least the target value of the second CGM metric based on the TPR and FPR of the classifier for each of the plurality of threshold targets of the second CGM metric comprises:

[0224] For each of the plurality of threshold targets measured by the second CGM, the receiver operating characteristic (ROC) curve of the classifier is determined based on the TPR and FPR of the classifier;

[0225] Select the first data point of the ROC curve, the first data point being closest to the point where TPR is 1 and FPR is zero; and

[0226] The threshold target associated with the first data point is identified as the target value of the second CGM metric corresponding to the target value of the first CGM metric.

[0227] Clause 18. The system according to any one of Clauses 15 to 17, wherein the first CGM metric comprises:

[0228] Glucose Management Index (GMI);

[0229] The percentage of time (TIR) ​​during which glucose levels range from 70 mg / dL to 180 mg / dL;

[0230] The percentage of time during which glucose levels were in the range of 70 mg / dL to 140 mg / dL (TITR);

[0231] The percentage of time during which glucose levels were above 180 mg / dL (TA180);

[0232] The percentage of time during which glucose levels were above 250 mg / dL (TA250);

[0233] The percentage of time during which glucose levels were below 70 mg / dL (TB70); or

[0234] The percentage of time during which glucose levels were below 54 mg / dL (TB54).

[0235] Clause 19. The system according to any one of Clauses 15 to 18, wherein the second CGM metric includes GMI, TIR, TITR, TA180, TA250, TB70, or TB54.

[0236] Clause 20. The system according to any one of Clauses 15 to 19, wherein said operation further comprises:

[0237] Determine whether the estimated target value of at least the second CGM metric meets predetermined criteria; and

[0238] Configure the insulin delivery system based on the target value of the first CGM metric.

Claims

1. A processor-implemented method, the processor-implemented method comprising: Receive the target value as input for the first continuous glucose monitoring (CGM) metric; Estimate the target value of at least a second CGM metric corresponding to the target value of the first CGM metric; and Provide the user with the estimated target value of at least the second CGM metric.

2. The processor-implemented method according to claim 1, wherein the first CGM metric includes: Glucose Management Index (GMI); The percentage of time (TIR) ​​during which glucose levels range from 70 mg / dL to 180 mg / dL; The percentage of time during which glucose levels were in the range of 70 mg / dL to 140 mg / dL (TITR); The percentage of time during which glucose levels were above 180 mg / dL (TA180); The percentage of time during which glucose levels were above 250 mg / dL (TA250); The percentage of time during which glucose levels were below 70 mg / dL (TB70); or The percentage of time during which glucose levels were below 54 mg / dL (TB54).

3. The processor-implemented method of claim 1, wherein the second CGM metric includes GMI, TIR, TITR, TA180, TA250, TB70, or TB54.

4. The processor-implemented method of claim 1, wherein estimating the target value of at least the second CGM metric corresponding to the target value of the first CGM metric comprises: The classifier is used to classify the CGM data samples that satisfy and do not satisfy the target value of the first CGM metric into samples that satisfy or do not satisfy the second CGM metric for each of a plurality of threshold targets. For each of the plurality of threshold targets quantified by the second CGM, the true positive rate (TPR) and false positive rate (FPR) of the classifier are determined; as well as For each of the plurality of threshold targets in the second CGM metric, at least the target value of the second CGM metric is identified based on the TPR and FPR of the classifier.

5. The processor-implemented method of claim 4, wherein identifying at least the target value of the second CGM metric based on the TPR and FPR of the classifier for each of the plurality of threshold targets of the second CGM metric comprises: For each of the plurality of threshold targets measured by the second CGM, the receiver operating characteristic (ROC) curve of the classifier is determined based on the TPR and FPR of the classifier; Select the first data point of the ROC curve, the first data point being closest to the point where TPR is 1 and FPR is zero; and The threshold target associated with the first data point is identified as the target value of the second CGM metric corresponding to the target value of the first CGM metric.

6. The processor-implemented method of claim 4, wherein the CGM data sample comprises at least one of the following: The user's CGM data sample; CGM data sample of the group; CGM data samples from users in age groups; CGM data sample from users of automated insulin delivery systems; CGM data samples from users with type 1 diabetes and / or users with type 2 diabetes; or Training samples and test samples.

7. The processor-implemented method of claim 4, wherein the classifier comprises a binary classifier.

8. The processor-implemented method according to claim 1, further comprising: Determine that the estimated target value of at least the second CGM metric satisfies the predetermined criteria; as well as Configure the insulin delivery system based on the target value of the first CGM metric.

9. A processor-implemented method, the processor-implemented method comprising: The classifier is used to classify continuous glucose monitoring (CGM) data samples that meet and do not meet the target value of the first CGM metric into samples that meet or do not meet the target of the second CGM metric. For each of the plurality of threshold targets quantified by the second CGM, the true positive rate (TPR) and false positive rate (FPR) of the classifier are determined; as well as For each of the plurality of threshold targets in the second CGM metric, at least the target value of the second CGM metric is identified based on the TPR and FPR of the classifier.

10. The processor-implemented method of claim 9, wherein identifying at least the target value of the second CGM metric based on the TPR and FPR of the classifier for each of the plurality of threshold targets of the second CGM metric comprises: For each of the plurality of threshold targets measured by the second CGM, the receiver operating characteristic (ROC) curve of the classifier is determined based on the TPR and FPR of the classifier; Select the first data point of the ROC curve, the first data point being closest to the point where TPR is 1 and FPR is zero; and The threshold target associated with the first data point is identified as the target value of the second CGM metric corresponding to the target value of the first CGM metric.

11. The processor-implemented method of claim 9, wherein the first CGM metric comprises: Glucose Management Index (GMI); The percentage of time (TIR) ​​during which glucose levels range from 70 mg / dL to 180 mg / dL; The percentage of time during which glucose levels were in the range of 70 mg / dL to 140 mg / dL (TITR); The percentage of time during which glucose levels were above 180 mg / dL (TA180); The percentage of time during which glucose levels were above 250 mg / dL (TA250); The percentage of time during which glucose levels were below 70 mg / dL (TB70); or The percentage of time during which glucose levels were below 54 mg / dL (TB54).

12. The processor-implemented method of claim 9, wherein the second CGM metric includes GMI, TIR, TITR, TA180, TA250, TB70, or TB54.

13. The processor-implemented method of claim 9, wherein the CGM data sample comprises at least one of the following: User's CGM data sample; CGM data sample of the group; CGM data samples from users in age groups; CGM data samples from users of automated insulin delivery systems; or CGM data sample of users with type 1 diabetes and / or users with type 2 diabetes.

14. The processor-implemented method of claim 9, wherein the CGM data sample includes CGM data samples from users of an automated insulin delivery system.

15. A system comprising: One or more processors; One or more processor-readable storage media storing instructions that, when executed by the one or more processors, cause the following operations: Receive the target value as input for the first continuous glucose monitoring (CGM) metric; Estimate the target value of at least a second CGM metric corresponding to the target value of the first CGM metric; and Provide the user with the estimated target value of at least the second CGM metric.

16. The system of claim 15, wherein estimating the target value of the second CGM metric that corresponds at least to the target value of the first CGM metric comprises: The classifier is used to classify the CGM data samples that satisfy and do not satisfy the target value of the first CGM metric into samples that satisfy or do not satisfy the second CGM metric for each of a plurality of threshold targets. For each of the plurality of threshold targets quantified by the second CGM, the true positive rate (TPR) and false positive rate (FPR) of the classifier are determined; as well as For each of the plurality of threshold targets in the second CGM metric, at least the target value of the second CGM metric is identified based on the TPR and FPR of the classifier.

17. The system of claim 16, wherein identifying at least the target value of the second CGM metric based on the TPR and FPR of the classifier for each of the plurality of threshold targets of the second CGM metric comprises: For each of the plurality of threshold targets measured by the second CGM, the receiver operating characteristic (ROC) curve of the classifier is determined based on the TPR and FPR of the classifier; Select the first data point of the ROC curve, the first data point being closest to the point where TPR is 1 and FPR is zero; and The threshold target associated with the first data point is identified as the target value of the second CGM metric corresponding to the target value of the first CGM metric.

18. The system of claim 15, wherein the first CGM metric comprises: Glucose Management Index (GMI); The percentage of time (TIR) ​​during which glucose levels range from 70 mg / dL to 180 mg / dL; The percentage of time during which glucose levels were in the range of 70 mg / dL to 140 mg / dL (TITR); The percentage of time during which glucose levels were above 180 mg / dL (TA180); The percentage of time during which glucose levels were above 250 mg / dL (TA250); The percentage of time during which glucose levels were below 70 mg / dL (TB70); or The percentage of time during which glucose levels were below 54 mg / dL (TB54).

19. The system of claim 15, wherein the second CGM metric includes GMI, TIR, TITR, TA180, TA250, TB70, or TB54.

20. The system of claim 15, wherein the operation further comprises: Determine whether the estimated target value of at least the second CGM metric meets predetermined criteria; as well as Configure the insulin delivery system based on the target value of the first CGM metric.

Images