Granted Patent B2

US 12,629,059 · App. 17/974,190 · Granted May 19, 2026

Inventors: Robert J. Dowd (San Diego, CA); Margaret A. Crawford (San Diego, CA); Mark Derdzinski (San Diego, CA); Lauren H. Jepson (San Diego, CA); Giada Acciaroli (San Diego, CA); Sarah Kate Pickus (San Diego, CA); Apurv U. Kamath (San Diego, CA)

Assignee: DEXCOM, INC.

A61B5/14532A61B5/7267A61B5/7275A61B5/742G16H50/30G16H50/70

Abstract

Claims (66)

1 . A method implemented in a continuous glucose level monitoring system, the method comprising:

obtaining, from a glucose sensor of the continuous glucose level monitoring system for each of multiple time periods, glucose measurements measured for a user for the time period, the glucose sensor being inserted at an insertion site of the user;

generating, based on the glucose measurements for each of the multiple time periods, a plurality of aggregate metrics for the user during the time period;

measuring, from the plurality of aggregate metrics, a first aggregate metric for the user during a first time period;

predicting, for the first time period, a predicted aggregate metric for the user based on the plurality of aggregate metrics generated for a series of multiple time periods preceding the first time period;

determining an error as an absolute difference between the predicted aggregate metric and the first aggregate metric for the first time period;

determining that a deviation is present when the error exceeds a pre-determined threshold;

generating, in response to determining that the error exceeds the pre-determined threshold, a user interface identifying the deviation; and

causing the user interface identifying the deviation to be displayed.

2 . The method of claim 1 , further comprising:

determining, for the first time period of the multiple time periods, whether the plurality of aggregate metrics for the first time period indicate an additional deviation by first glucose measurements measured for the first time period from glucose measurements measured during the series of multiple time periods preceding the first time period;

generating, in response to determining that the plurality of aggregate metrics for the first time period indicate the additional deviation, a user interface identifying the additional deviation; and

causing the user interface identifying the additional deviation to be displayed.

3 . The method of claim 1 , wherein the plurality of aggregate metrics includes high blood glucose index values.

4 . The method of claim 1 , wherein the plurality of aggregate metrics includes low blood glucose index values.

5 . The method of claim 1 , wherein the determining includes determining whether the plurality of aggregate metrics for the first time period indicate the deviation using a machine learning system trained with sets of multiple aggregate metrics as training data and trained to minimize a loss between the predicted aggregate metric and an actual aggregate metric.

6 . The method of claim 1 , wherein each of the multiple time periods comprises 30 minutes and the multiple time periods include 24 time periods.

7 . The method of claim 1 , further comprising:

obtaining, from the glucose sensor for a time period of a current day, glucose measurements measured for the user for the time period;

generating, based on the glucose measurements for the time period, a second aggregate metric for the user during the time period of the current day;

generating, based on glucose measurements for corresponding time periods on each of multiple preceding days, aggregate metrics for the user during the corresponding time periods on the multiple preceding days;

determining whether the second aggregate metric indicates an additional deviation by first glucose measurements measured for the first time period of the current day from glucose measurements measured during the corresponding time periods on the multiple preceding days;

generating, in response to determining that the second aggregate metric indicates the additional deviation, a user interface identifying the additional deviation; and

causing the user interface identifying the additional deviation to be displayed.

8 . The method of claim 1 , wherein the deviation is one of multiple deviations, and the method further comprising:

identifying a population of which the user is a part; and

selecting one of the multiple deviations based on the population, the population being one of multiple different populations of users, and the generating comprising generating the user interface identifying the selected deviation.

9 . The method of claim 8 , wherein the identifying the population of which the user is a part comprises identifying the population of which the user is a part based on an age of the user or a diabetes diagnosis of the user.

10 . A computing device comprising:

a processor;

a display device; and

computer-readable storage media having stored thereon multiple instructions of an application that, responsive to execution by the processor, cause the processor to:

obtain, from a glucose sensor of a continuous glucose level monitoring system for each of multiple time periods, glucose measurements measured for a user for the time period, the glucose sensor being inserted at an insertion site of the user;

generate, based on the glucose measurements for each of the multiple time periods, plurality of aggregate metrics for the user during the time period;

measure, from the plurality of aggregate metrics, a first aggregate metric for the user during a first time period;

predict, for the first time period, a predicted aggregate metric for the user based on the plurality of aggregate metrics generated for a series of multiple time periods preceding the first time period;

determine an error as an absolute difference between the predicted aggregate metric and the first aggregate metric for the first time period;

determine that a deviation is present when the error exceeds a pre-determined threshold;

generate, in response to determining that the error exceeds the pre-determined threshold, a user interface identifying the deviation; and

cause the user interface identifying the deviation to be displayed on the display device.

11 . The computing device of claim 10 , wherein the deviation is one of multiple deviations, and the multiple instructions further cause the processor to:

identify a population of which the user is a part; and

select one of the multiple deviations based on the population, the population being one of multiple different populations of users, and wherein to generate the user interface is to generate the user interface identifying the selected deviation.

12 . The computing device of claim 10 , wherein the multiple instructions further cause the processor to:

determine, for the first time period of the multiple time periods, whether the plurality of aggregate metrics for the first time period indicate an additional deviation by first glucose measurements measured for the first time period from glucose measurements measured during the series of multiple time periods preceding the first time period;

generate, in response to determining that the plurality of aggregate metrics for the first time period indicate the additional deviation, a user interface identifying the additional deviation; and

cause the user interface identifying the additional deviation to be displayed on the display device.

13 . The computing device of claim 10 , wherein to determine whether the first aggregate metric for the first time period indicates the deviation is to determine whether the plurality of aggregate metrics for the first time period indicate the deviation using a machine learning system trained with sets of multiple aggregate metrics as training data and trained to minimize a loss between a predicted risk aggregate metric and an actual aggregate metric.

14 . The computing device of claim 13 , wherein the user is part of one population of multiple different populations of users, and the using a machine learning system includes using a machine learning system trained with training data for the one population.

15 . A method implemented in a continuous glucose level monitoring system, the method comprising:

obtaining, from a glucose sensor of the continuous glucose level monitoring system for a time period of a current day, glucose measurements measured for a user for the time period, the glucose sensor being inserted at an insertion site of the user;

generating, based on the glucose measurements for the time period, an aggregate metric for the user during the time period of the current day;

generating, based on glucose measurements for corresponding time periods on each of multiple preceding days, a plurality of aggregate metrics for the user during the corresponding time periods on the multiple preceding days;

predicting, for the time period of the current day, a predicted aggregate metric for the user based on the plurality of aggregate metrics generated for the user during the corresponding time periods on the multiple preceding days;

determining an error as an absolute difference between the predicted aggregate metric and the aggregate metric for the user during the time period of the current day;

determining that a deviation is present when the error exceeds a pre-determined threshold;

generating, in response to determining that the error exceeds the pre-determined threshold, a user interface identifying the deviation; and

causing the user interface identifying the deviation to be displayed.

16 . The method of claim 15 , further comprising:

generating a value representing the plurality of aggregate metrics for the user during the corresponding time periods on the multiple preceding days, and

wherein the determining comprises determining that the aggregate metric for the time period of the current day indicates an additional deviation by the glucose measurements measured for the time period of the current day from glucose measurements measured during the corresponding time periods on the multiple preceding days in response to a difference between the value representing the plurality of aggregate metrics for the user during the corresponding time periods on the multiple preceding days and the aggregate metric for the user during the time period of the current day exceeding a threshold amount.

17 . The method of claim 16 , wherein the value representing the plurality of aggregate metrics for the user during the corresponding time periods on the multiple preceding days includes a mean or average of the plurality of aggregate metrics for the user during the corresponding time periods on the multiple preceding days.

18 . The method of claim 15 , wherein the deviation is one of multiple deviations, and the method further comprising:

identifying a population of which the user is a part; and

selecting one of the multiple deviations based on the population, wherein the population is one of multiple different populations of users, and the generating the user interface comprises generating the user interface identifying the selected deviation.

19 . The method of claim 18 , wherein the identifying the population of which the user is a part comprises identifying the population of which the user is a part based on an age of the user or a diabetes diagnosis of the user.

Assignments (1)

Continuity (3)

References Cited (34)