US-12620469-B2 - Therapeutic zone assessor

Abstract

Systems and methods are provided for identifying therapeutic zones where there is glycemic dysfunction of a specific type that can be addressed by making strategic changes to behavior and/or therapy parameters. Systems and methods described herein evaluate large historical data sets to: identify a therapeutic zone or zones with glycemic dysfunction that are most readily addressable; quantify the glycemic impact of a plurality of different therapeutic adjustments in terms of either adjustments to historical doses or the parameters of a prospective dosing strategy to determine the highest possible improvement; and/or identify patient dosing strategies to provide therapy recommendations adapted for the patient's preferred behavioral dosing strategy.

Inventors

• Stephen D. Patek

Assignees

• DEXCOM, INC.

Dates

Publication Date

20260505

Application Date

20201216

Claims (20)

1. 1 . A method for reducing glycemic dysfunction, comprising: receiving, via an interface circuit, continuous glucose monitoring (CGM) data and corresponding insulin delivery data; automatically, with at least one processor executing stored instructions, identifying a therapeutic improvement opportunity by retrospectively analyzing a historical glucose dataset generated from the received glucose data, the therapeutic improvement opportunity occurring during one or more time zones where there is glycemic dysfunction; determining, by the processor, a plurality of candidate changes to at least one insulin delivery parameter for the one or more time zones that have been identified; executing a replay simulation that applies each candidate change to the historical glucose dataset to generate a respective simulated glycemic-risk profile; computing, for each simulated glycemic-risk profile, a quantitative risk-reduction metric and selecting, by the processor, at least one candidate change that provides a largest reduction in the risk-reduction metric; generating, by the processor, pump-specific, machine-readable control instructions that encode the selected at least one candidate change; outputting the control instructions via a wired or wireless transceiver operably coupled to an insulin infusion pump; and administering the insulin therapy by automatically adjusting parameters and/or timing of insulin therapy in the insulin infusion pump in real time in response to the transmitted control instructions, thereby implementing the at least one candidate change that is outputted.
2. 2 . The method of claim 1 , wherein the glucose and insulin data is received from at least one of a patient or a connected system or device, and wherein multiple simulated glycemic-risk profiles are associated with the one or more time zones, and wherein implementing the at least one candidate change addresses multiple correlated glycemic risks in the one or more time zones.
3. 3 . The method of claim 1 , wherein identifying the therapeutic improvement opportunity comprises receiving a user selection identifying a specific insulin therapy or time of day to be optimized, wherein the user selection is at least one of a mealtime, a time of day, or a parameter setting.
4. 4 . The method of claim 3 , wherein the parameter setting is a carb ratio.
5. 5 . The method of claim 1 , wherein the candidate changes to insulin therapy comprise percentage increases or decreases to bolus therapy or basal therapy.
6. 6 . The method of claim 1 , wherein the candidate changes to insulin therapy comprise changes to insulin delivery parameters associated with bolus therapy or basal therapy.
7. 7 . The method of claim 1 , wherein the candidate changes are in terms of carb ratios, correction factors, basal rates, or profiles.
8. 8 . The method of claim 1 , wherein the candidate changes comprise basal dose sensitivity.
9. 9 . The method of claim 1 , wherein the candidate changes comprise percentage change to basal or bolus doses in therapeutic zones.
10. 10 . The method of claim 1 , wherein quantifying the improvement of the candidate changes comprises comparing risk profile values.
11. 11 . The method of claim 1 , wherein outputting at least one of the candidate changes based on the improvement comprises outputting the candidate change that provides the optimized risk profile.
12. 12 . The method of claim 1 , wherein outputting at least one of the candidate changes comprises providing an output in the form of a graph illustrating at least one of a candidate change or an optimized risk output to a user interface or connected device.
13. 13 . The method of claim 12 , wherein the connected device comprises a bolus calculator.
14. 14 . The method of claim 12 , wherein the output is provided by a natural language processor to describe a candidate change and an optimized risk outcome.
15. 15 . The method of claim 12 , wherein the output identifies which therapeutic zones or zone groups have been optimized.
16. 16 . The method of claim 1 wherein the historical glucose dataset is collected over a period of one week.
17. 17 . A system comprising: at least one processor; and a non-transitory computer readable medium comprising instructions that, when executed by the at least one processor, cause the system to: receive, via an interface circuit, continuous glucose monitoring (CGM) data and corresponding insulin delivery data; automatically, with at least one processor executing stored instructions, identify a therapeutic improvement opportunity by retrospectively analyzing a historical glucose dataset generated from the received glucose data, the therapeutic improvement opportunity occurring during one or more time zones where there is glycemic dysfunction; determine, by the processor, a plurality of candidate changes to at least one insulin delivery parameter for the one or more time zones that have been identified; execute a replay simulation that applies each candidate change to the historical glucose dataset to generate a respective simulated glycemic-risk profile; compute, for each simulated glycemic-risk profile, a quantitative risk-reduction metric and select, by the processor, at least one candidate change that provides a largest reduction in the risk-reduction metric; generate, by the processor, pump-specific, machine-readable control instructions that encode the selected at least one candidate change; output the control instructions via a wired or wireless transceiver operably coupled to an insulin infusion pump; and administer the insulin therapy by automatically adjusting parameters and/or timing of insulin therapy in the insulin infusion pump in real time in response to the transmitted control instructions, thereby implementing the at least one candidate change that is outputted.
18. 18 . The system of claim 17 , wherein the glucose and insulin data is received from at least one of a patient or a connected system or device.
19. 19 . The system of claim 17 , wherein identifying the therapeutic improvement opportunity comprises receiving a user selection of at least one of a mealtime, a time of day, or a parameter setting.
20. 20 . The system of claim 19 , wherein the parameter setting is a carb ratio.

Description

INCORPORATION BY REFERENCE TO RELATED APPLICATION This application claims the benefit of priority to U.S. Provisional Patent Application No. 62/950,029, filed on Dec. 18, 2019, entitled “THERAPEUTIC ZONE ASSESSOR,” the contents of which are hereby incorporated by reference in its entirety, and is hereby expressly made a part of this specification. BACKGROUND With the growing adoption of CGM (continuous glucose monitoring) and connected devices, the availability and reliability of glucose time-series data has increased in recent years. Identifying multiple patterns in large historic data sets requires a level of complexity that cannot be addressed by human evaluation due at least in part to overlapping symptoms in those patterns. Even with pattern analysis tools, doctors cannot reliably determine what aspect of diabetes therapy is most readily addressable for each unique patient situation based on a review of the data. Because of the numerous variables and factors involved in diabetes management, current practices for identifying patterns and making recommendations lack a reliability and ease of use that considers things such as credibility of data, therapeutic addressability of certain aspects of diabetes risks, impact improvements, patient-preferred diabetes management strategies, and synthesis of related risks and therapies, for example. Often times, clinicians review CGM traces for a patient over a period of time, such as 14 days, and corresponding insulin delivery patterns, or review the data in a more consolidated format such as a plot that shows each of the 14 days of data overlaid on a 24 hour timeline in an attempt to visually highlight areas of stronger patterns within a particular time of day. However, visualization of the data cannot easily highlight many of the important factors, risks, and potential outcomes needed for effective therapy optimization. Additionally, the credibility of the data is unknown or unclear from visual inspection. It is with respect to these and other considerations that the various aspects and embodiments of the present disclosure are presented. SUMMARY Systems and methods are provided for identifying therapeutic zones where there is glycemic dysfunction of a specific type that can be addressed by making strategic changes to behavior and/or therapy parameters. Systems and methods described herein evaluate large historical data sets to: identify a therapeutic zone or zones with glycemic dysfunction that are most readily addressable; quantify the glycemic impact of a plurality of different therapeutic adjustments in terms of either adjustments to historical doses or the parameters of a prospective dosing strategy to determine the highest possible improvement; and/or identify patient dosing strategies to provide therapy recommendations adapted for the patient's preferred behavioral dosing strategy. In an implementation, a method comprises: deriving at least one single symptom-specific risk profile, using a glycemic risk profiler; determining, using a therapeutic zone assessor, at least one therapeutically correlated zone associated with the at least one single symptom-specific risk profile; determining, using a zone importance quantifier, an importance value of the at least one therapeutically correlated zone; and outputting information based on the importance value. Implementations may include some or all of the following features. The method further comprises receiving glucose data, wherein deriving the at least one single symptom-specific risk profile uses the glucose data. The glucose data comprises at least one of CGM (continuous glucose monitoring) readings, confidence readings assigned to CGM values, self-monitoring blood glucose readings, or retrospectively calibrated or corrected CGM readings. The glucose data encompasses a time period of at least one week. The at least one single symptom-specific risk profile describes either hypoglycemic risk or hyperglycemic risk as a function of the time of day using glucose data. Deriving the at least one single symptom-specific risk profile comprises at least one of evaluating steepness (first and second order derivatives of a curve), frequency, severity, curvature, average value of profile across 24 hours, or variability of the profile (mean and standard deviation). The at least one single symptom-specific risk profile is indicative of glycemic dysfunction based the CGM signal over a selected time period, indicating recurring windows of time characterized by a predefined severity and frequency of hypoglycemia or hyperglycemia over the selected time period. The at least one single symptom-specific risk profile represents at least one of hypoglycemia isolated from hyperglycemia, or hyperglycemia isolated from hypoglycemia. Determining the at least one therapeutically correlated zone comprises identifying the at least one therapeutically correlated zone from the at least one single symptom-specific risk profile. The at least on