IP Library Granted Patent US 12,226,612
Granted Patent B2
US 12,226,612 · App. 17/715,589 · Granted Feb 18, 2025

Compensation for missing readings from a glucose monitor in an automated insulin delivery system

Inventors: Yibin Zheng (Hartland, WI); Joon Bok Lee (Acton, MA); Ashutosh Zade (San Diego, CA); Jason O'Connor (Acton, MA)
Assignee: INSULET CORPORATION
A61M5/1723G06F17/17A61M2205/3303A61M2205/3331A61M2205/3507A61M2205/3553A61M2205/52A61M2230/201
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,226,612
App. No.
17/715,589
Granted
Feb 18, 2025
Kind
B2
Abstract

Exemplary embodiments may address the problem of missing blood glucose concentration readings from a glucose monitor that transmits blood glucose concentration readings over a wireless connection due to problems with the wireless connection. In the exemplary embodiments, an automated insulin delivery (AID) device uses an estimate in place of a missing blood glucose concentration reading in determining a predicted future blood glucose concentration reading for a user. Thus, the AID device is able to operate normally in generating insulin delivery settings despite not receiving a current blood glucose concentration reading for a current cycle. There is no need to suspend delivery of insulin to the user due to the missing blood glucose concentration reading.

Claims (39)

1. An electronic device, comprising:

a wireless interface with a glucose sensor for providing glucose level readings of a user;

a storage medium storing programming instructions; and

a processor for executing the programming instructions in the storage medium to:

where a current glucose level reading for the user is successfully received over the wireless interface from the glucose sensor for a current control cycle, set insulin delivery settings for delivery of the insulin to the user from an automated insulin delivery (AID) device for the current control cycle based on predicted future glucose level readings for the user, wherein the predicted future glucose level readings are based on, at least in part, glucose level readings from the glucose sensor from previous control cycles; and

where the current glucose level reading for the user is not successfully received over the wireless interface from the glucose sensor for the current control cycle, estimate the glucose level reading that was not successfully received over the wireless interface from the glucose sensor, and set the insulin delivery settings for delivery of the insulin to the user from the AID device for the current control cycle based on the predicted future glucose level readings for the user, wherein the predicted future glucose level readings are based on the glucose level readings from the glucose sensor from previous control cycles and the estimate of the current glucose level reading that was not successfully received over the wireless interface from the glucose sensor.

2. The electronic device of claim 1 , wherein the estimate of the current glucose level reading is one of the predicted future glucose level readings.

3. The electronic device of claim 1 , wherein the estimate of the current glucose level reading is determined by applying interpolation of glucose level readings from the glucose sensor from previous control cycles.

4. The electronic device of claim 3 , wherein the estimate of the of the current glucose level reading is determined by summing a most recent received glucose level reading with an average change between most recent ones of the glucose level readings.

5. The electronic device of claim 1 , wherein the estimate of the current glucose level reading is a most recently received glucose level reading.

6. A method performed by a processor, comprising:

where a current glucose level reading for a user is successfully received by an automated insulin delivery (AID) device over a wireless interface from the glucose sensor for a current control cycle, setting insulin delivery settings for delivery of the insulin by the AID device to the user of the AID device for the current control cycle based on predicted future glucose levels for the user, wherein the predicted future glucose levels are based, at least in part, on glucose level readings from the glucose sensor for previous control cycles; and

where the glucose level reading for the user is not successfully received by the AID device over the wireless interface from the glucose sensor for the current control cycle, estimating the current glucose level reading that was not successfully received over the wireless interface from the glucose sensor, and setting the insulin delivery settings for delivery of the insulin to the user for the current control cycle based on the predicted future glucose levels for the user, wherein the predicted future glucose levels are based on the glucose level readings from the glucose sensor from previous control cycles and the estimate of the current glucose level reading that was not successfully received over the wireless interface from the glucose sensor.

7. The method of claim 6 , wherein the estimate of the current glucose level reading is one of the predicted future glucose levels.

8. The method of claim 6 , wherein the estimate of the current glucose level reading is determined by applying interpolation of glucose level readings from the glucose sensor for previous control cycles.

9. The method of claim 8 , wherein the estimate of the of the current glucose level reading is determined by summing a most recently received glucose level reading with an average change between most recent ones of the glucose level readings received from the glucose sensor.

10. The method of claim 9 , wherein the estimate of the of the current glucose level reading is determined by summing a most recently received glucose level reading with an average change between two most recent ones of the glucose level readings received from the glucose sensor.

11. The method of claim 6 , wherein the estimate of the current blood glucose level reading is a most recently received glucose level reading.

12. A method performed by an automated insulin delivery (AID) device, comprising:

receiving glucose level readings for a user from a glucose sensor over a wireless interface at the AID device for control cycles;

where, for a given control cycle, a glucose level reading for the user is not received at the AID device from the glucose sensor over the wireless interface, determining an estimate of the glucose level reading for the user for the given control cycle;

using the received blood glucose level readings and the determined estimate for the given control cycle to predict a future glucose level reading;

setting insulin delivery settings of the AID device based on the predicted future glucose level for the user; and

when communication over the wireless interface with the glucose sensor is reestablished, replacing the estimate of the glucose level reading for the user for the given control cycle with a glucose level reading for the user from the glucose monitor for the given cycle in determining an estimate of the glucose level reading for the user for a next control cycle.

13. The method of claim 12 , wherein the determining the estimate of the glucose level reading for the user for the given control cycle comprises determining a trend of glucose level values from the received glucose level readings and using the trend to determine the estimate.

14. The method of claim 13 , wherein the using the trend of glucose level values comprises performing extrapolation based on the trend to obtain the estimate or performing interpolation based on the trend to obtain the estimate.

15. The method of claim 12 , further comprising using the using the glucose level reading for the given cycle in setting the insulin delivery settings of the AID device.

16. A non-transitory computer-readable storage medium storing instructions that when executed by a processor cause the processor to:

receive glucose level readings for a user from a glucose sensor over a wireless interface at the AID device for control cycles, wherein during normal operation a glucose level reading is received for each of the control cycles;

where, for a given control cycle, a glucose level reading for the user is not received at the AID device from the glucose sensor over the wireless interface, determine an estimate of the glucose level reading for the user for the given control cycle;

use the received glucose level readings and the determined estimate for the given control cycle to predict a future glucose level reading;

set insulin delivery settings of the AID device based on the predicted future glucose level for the user; and

when communication over the wireless interface with the glucose sensor is reestablished, replace the estimate of the glucose level reading for the user for the given control cycle with a glucose level reading for the user from the glucose monitor for the given cycle in determining an estimate of the glucose level reading for the user for a next control cycle.

17. The non-transitory computer-readable storage medium of claim 16 , wherein the determining an estimate of the glucose level reading for the user for the given control cycle comprises determining a trend of glucose level values from the received glucose level readings and using the trend to determine the estimate.

18. The non-transitory computer-readable storage medium of claim 16 , wherein the using the trend comprises performing extrapolation based on the trend to obtain the estimate or performing interpolation based on the trend to obtain the estimate.

19. A non-transitory computer-readable storage medium storing instructions that when executed by a processor cause the processor to:

where a current glucose level reading for a user is successfully received by an automated insulin delivery (AID) device over a wireless interface from the glucose sensor for a current control cycle, set insulin delivery settings for delivery of the insulin by the AID device to the user for the current control cycle based on predicted future glucose levels for the user, wherein the predicted future glucose levels are based, at least in part, on glucose level readings from the glucose sensor for previous control cycles; and

where the glucose level reading for the user is not successfully received by the AID device over the wireless interface from the glucose sensor for the current control cycle, estimate the current glucose level reading that was not successfully received over the wireless interface from the glucose sensor, and set the insulin delivery settings for delivery of the insulin to the user for the current control cycle based on the predicted future glucose levels for the user, wherein the predicted future glucose levels are based on the glucose level readings from the glucose sensor from previous control cycles and an estimate of the current glucose level reading that was not successfully received over the wireless interface from the glucose sensor.

20. The non-transitory computer-readable storage medium of claim 19 , wherein the estimate of the current glucose level reading is one of the predicted future glucose levels for the current control cycle or determined by applying interpolation of blood glucose level readings from the glucose sensor from previous control cycles.

Assignments (2)
SECURITY AGREEMENT SUPPLEMENT Recorded Dec 1, 2022
From: INSULET CORPORATION
To: MORGAN STANLEY SENIOR FUNDING, INC., AS COLLATERAL AGENT
Reel/Frame 062035/0785 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 7, 2022
From: ZHENG, YIBIN; LEE, JOON BOK; ZADE, ASHUTOSH; O'CONNOR, JASON
To: INSULET CORPORATION
Reel/Frame 059535/0344 →
Continuity (2)
Continuation 16791648 · Feb 14, 2020
Related Publication 20220226576A1 · Jul 21, 2022
References Cited (69)
US 5007286A · Malcolm et al. · 1991 [cited by applicant]
US 6553841B1 · Blouch · 2003 [cited by applicant]
US 9233204B2 · Booth et al. · 2016 [cited by applicant]
US 10583250B2 · Mazlish et al. · 2020 [cited by applicant]
US 10737024B2 · Schmid · 2020 [cited by applicant]
US 10987468B2 · Mazlish et al. · 2021 [cited by applicant]
US 11197964B2 · Sjolund et al. · 2021 [cited by applicant]
US 11260169B2 · Estes · 2022 [cited by applicant]
US 11324889B2 · Zheng · 2022 [cited by examiner]
US 20080172026A1 · Blomquist · 2008 [cited by applicant]
US 20090054753A1 · Robinson et al. · 2009 [cited by applicant]
US 20110124996A1 · Reinke et al. · 2011 [cited by applicant]
US 20140146202A1 · Boss et al. · 2014 [cited by applicant]
US 20150120317A1 · Mayou et al. · 2015 [cited by applicant]
US 20150134265A1 · Kohlbrecher et al. · 2015 [cited by applicant]
US 20150359490A1 · Massey et al. · 2015 [cited by applicant]
US 20150366945A1 · Greene · 2015 [cited by applicant]
US 20160038673A1 · Morales · 2016 [cited by applicant]
US 20160354543A1 · Cinar et al. · 2016 [cited by applicant]
US 20170049386A1 · Abraham et al. · 2017 [cited by applicant]
US 20170348482A1 · Duke et al. · 2017 [cited by applicant]
US 20180169334A1 · Grosman et al. · 2018 [cited by applicant]
US 20180200434A1 · Mazlish et al. · 2018 [cited by applicant]
US 20180200438A1 · Mazlish et al. · 2018 [cited by applicant]
US 20180289891A1 · Finan et al. · 2018 [cited by applicant]
US 20190076600A1 · Grosman et al. · 2019 [cited by applicant]
US 20200046268A1 · Patek et al. · 2020 [cited by applicant]
US 20200342974A1 · Chen et al. · 2020 [cited by applicant]
US 20210098105A1 · Lee et al. · 2021 [cited by applicant]
US 20220023536A1 · Graham et al. · 2022 [cited by applicant]
AU 2015200834A1 · 2015 [cited by applicant]
EP 2897071B1 · 2019 [cited by applicant]
JP 2002085556A · 2002 [cited by applicant]
JP 2019525276A · 2019 [cited by applicant]
WO 2005082436A1 · 2005 [cited by applicant]
WO 2007078937A1 · 2007 [cited by applicant]
WO 2009134380A2 · 2009 [cited by applicant]
WO 2010053702A1 · 2010 [cited by applicant]
WO 2010132077A1 · 2010 [cited by applicant]
WO 2010138848A1 · 2010 [cited by applicant]
WO 2013177565A1 · 2013 [cited by applicant]
WO 2014123998A2 · 2014 [cited by applicant]
WO 2015187366A1 · 2015 [cited by applicant]
WO 2016004088A1 · 2016 [cited by applicant]
WO 2017091624A1 · 2017 [cited by applicant]
WO 2018067748A1 · 2018 [cited by applicant]
WO 2018136799A1 · 2018 [cited by applicant]
WO 2019213493A1 · 2019 [cited by applicant]
WO 2019246381A1 · 2019 [cited by applicant]
WO 2021011738A1 · 2021 [cited by applicant]
International Search Report and Written Opinion for the International Patent Application No. PCT/US2022/013470, mailed May 6, 2022, 14 pages. [cited by applicant]
International Search Report and Written Opinion for the International Patent Application No. PCT/US2022/013473, mailed May 6, 2022, 13 pages. [cited by applicant]
International Search Report and Written Opinion for the International Patent Application No. PCT/US2022/019079, mailed Jun. 2, 2022, 14 pages. [cited by applicant]
International Search Report and Written Opinion for the International Patent Application No. PCT/US2022/018453, mailed Jun. 2, 2022, 13 pages. [cited by applicant]
International Search Report and Written Opinion for the International Patent Application No. PCT/US22/018700, mailed Jun. 7, 2022, 13 pages. [cited by applicant]
International Search Report and Written Opinion for the International Patent Application No. PCT/US22/019080, mailed Jun. 7, 2022, 14 pages. [cited by applicant]
International Search Report and Written Opinion for the International Patent Application No. PCT/US22/019664, mailed Jun. 7, 2022, 14 pages. [cited by applicant]
International Search Report and Written Opinion for the International Patent Application No. PCT/US21/060618, mailed Mar. 21, 2022, 15 pages. [cited by applicant]
Herrero Pau et al: “Enhancing automatic closed-loop glucose control in type 1 diabetes with an adaptive meal bolus calculator—in silicoevaluation under intra-day variability”, Computer Methods and Programs in Biomedicin… [cited by applicant]
Marie Aude Qemerais: “Preliminary Evaluation of a New Semi-Closed-Loop Insulin Therapy System over the prandial beriod in Adult Patients with type I diabetes: the WP6. 0 Diabeloop Study”, Journal of Diabetes Science and… [cited by applicant]
Anonymous: “Kernel density estimation”, Wikipedia, Nov. 13, 2020 (Nov. 13, 2020), pp. 1-12, XP055895569, Retrieved from the Internet: URL:https://en.wikipedia.org/w/index.php?title=Kernel_density_estimation&oldid=988508… [cited by applicant]
Anonymous: “openaps / oref0 /lib/determine-basal-js”, openaps repository, Nov. 9, 2019 (Nov. 9, 2019), pp. 1-17, XP055900283, Retrieved from the Internet: URL:https://github.com/openaps/oref0/blob/master/lib/determine-b… [cited by applicant]
Anonymous: “AndroidAPS screens”, AndroidAPS documentation, Oct. 4, 2020 (Oct. 4, 2020), pp. 1-12, XP055894824, Retrieved from the Internet: URL:https://github.com/openaps/AndroidAPSdocs/blob/25d8acf8b28262b411b34f416f17… [cited by applicant]
Kozak Milos et al: “Issue #2473 of AndroidAPS”, MilosKozak / AndroidAPS Public repository, Mar. 4, 2020 (Mar. 4, 2020), pp. 1-4, XP055900328, Retrieved from the Internet: URL:https://github.com/MilosKozak/AndroidAPS/iss… [cited by applicant]
Medication Bar Code System Implementation Planning Section I: A Bar Code Primer for Leaders, Aug. 2013. [cited by applicant]
Medication Bar Code System Implementation Planning Section II: Building the Case for Automated Identification of Medications, Aug. 2013. [cited by applicant]
Villareal et al. (2009) in: Distr. Comp. Art. Intell. Bioninf. Soft Comp. Amb. Ass. Living; Int. Work Conf. Art. Neural Networks (IWANN) 2009, Lect. Notes Comp. Sci. vol. 5518; S. Omatu et al. (Eds.), pp. 870-877. [cited by applicant]
Fox, Ian G .; Machine Learning for Physiological Time Series: Representing and Controlling Blood Glucose for Diabetes Management; University of Michigan. ProQuest Dissertations Publishing, 2020. 28240142. (Year: 2020). [cited by applicant]
International Search Report and Written Opinion for the International Patent Application No. PCT/US2022/012896, mailed Apr. 22, 2022, 15 pages. [cited by applicant]
Cited By (1)
US 12,673,160