Method for modification of insulin delivery during pregnancy in automatic insulin delivery systems
The disclosed embodiments are directed to methods for dynamically adjusting the total daily insulin requirements of a user during pregnancy, based on the gestational week. An initial estimate of the adjusted total daily insulin requirement may be calculated as a multiple of the pre-pregnancy total daily insulin requirement, based on an average scale factor from a population of pregnant women suffering from Type I diabetes mellitus. An automatic drug delivery device may adjust the initial estimate of the total daily insulin requirement based on blood glucose level readings from a continuous glucose monitor during the course of the pregnancy.
1 . A method for modifying an insulin requirement for a user during pregnancy, comprising:
determining a pre-pregnancy total daily insulin requirement of the user;
determining progress of the pregnancy cycle;
scaling the pre-pregnancy total daily insulin requirement by a factor as a function of the progress of the pregnancy cycle to obtain a scaled total daily insulin requirement;
obtaining a plurality of blood glucose readings from the user;
determining that the scaled total daily insulin requirement does not effectively control the blood glucose level of the user, based on a predetermined percent of the blood glucose readings of the user being outside of a desirable range; and
determining an adapted total daily insulin requirement by adjusting the scaled total daily insulin requirement up or down based on the determination that the scaled total daily insulin requirement does not effectively control the blood glucose level of the user.
2 . The method of claim 1 further comprising:
determining, for a predetermined period of time, a percentage of the scaled total daily insulin requirement above or below a level necessary to effectively control the blood glucose level of the user; and
adjusting the scaled total daily insulin requirement for a next predetermined period by the percentage or a portion of the percentage to obtain the adapted total daily insulin requirement.
3 . The method of claim 1 wherein a first predetermined percentage of the adapted total daily insulin requirement of the user is delivered as one or more basal doses.
4 . The method of claim 3 further comprising:
automatically detecting an ingestion of a meal by the user; and
administering a bolus dose of insulin in response to the automatic detection.
5 . The method of claim 4 wherein the bolus dose is a second predetermined percentage of the adapted total daily insulin requirement of the user.
6 . The method of claim 5 further comprising:
tracking, for the predetermined period, a split of the adapted total daily insulin requirement for the predetermined period between basal doses and bolus doses; and
adjusting the first predetermined percentage of the adapted total daily insulin requirement of the user that is delivered as basal doses during a next predetermined period such that the total insulin administered to the user during the course of a day is the adapted total daily insulin requirement.
7 . The method of claim 4 wherein the bolus dose is delivered as a single dose.
8 . The method of claim 4 wherein the bolus dose is delivered as a dual dose wherein a first portion of the bolus dose is delivered upon detection of the ingestion of the meal and a second portion is delivered if an ingestion of a meal is detected after a predetermined period of time.
9 . The method of claim 4 wherein the bolus dose is delivered as multiple doses wherein a first portion of the bolus dose is delivered upon detection of the ingestion of the meal and a subsequent portions are delivered if an ingestion of a meal is detected after multiple predetermined periods of time.
10 . The method of claim 9 wherein a total of the bolus portions of the bolus dose administered do not exceed the second predetermined percentage.
11 . The method of claim 4 wherein the basal doses and bolus doses are delivered by an automatic drug delivery system.
12 . The method of claim 4 wherein the automatic detection of the ingestion of a meal further comprises:
receiving, from a sensor, periodic readings of the blood glucose levels of the user;
constructing a blood glucose curve for a moving window comprising a predetermined number of the blood glucose readings;
extracting, from the blood glucose curve, one or more features; and
determining, based on the extracted features, that a meal has been ingested by the user.
13 . The method of claim 12 wherein the extracted features are used in one or more decision trees.
14 . The method of claim 12 wherein the extracted features are input to a machine learning algorithm and wherein the machine learning algorithm outputs a meal or no-meal decision.
15 . The method of claim 14 wherein the one or more decision trees have varying delays built in such that meals are detected by each decision tree after the respective delay of the decision tree, the delay indicating a period of time after the blood glucose curve shows an initial rise in the blood glucose levels of the user.
16 . The method of claim 15 wherein the two or more of the decision trees having different delays are aggregated to form a cascaded model.
17 . The method of claim 16 wherein the feedback comprises information regarding food intake and exercise.
18 . The method of claim 12 wherein the extracted features include one or more of: one or more averages of portions of the blood glucose curve, one or more ranges of portions of the blood glucose curve, one or more coarse slopes of portions of the blood glucose curve, one or more slopes of portions of the blood glucose curves and one or more second derivatives of portions of the blood glucose curve.
19 . The method of claim 1 further comprising:
providing feedback to the user when the adapted total daily insulin requirement deviates from the scaled total daily insulin requirement.
20 . A method for modifying an insulin requirement for a user during pregnancy, comprising:
determining a pre-pregnancy total daily insulin requirement of the user;
determining progress of the pregnancy cycle;
scaling the pre-pregnancy total daily insulin requirement by a factor as a function of the progress of the pregnancy cycle to obtain a scaled total daily insulin requirement;
determining an adapted total daily insulin requirement by adjusting the scaled total daily insulin requirement up or down based on the determination that the scaled total daily insulin requirement does not effectively control the blood glucose level of the user;
accessing a catalog of meals ingested by the user, the catalog including quantities and macronutrient profile of the meals; and
recommending modifications to meals ingested by the user based on the deviance of the adapted total daily insulin requirement from the total daily insulin requirement.