US 11,887,715 B2
Dynamic equivalent on board estimator
Stephen D. Patek, Charlottesville, VA (US); and Enrique Campos-Náñez, San Diego, CA (US)
Assigned to DEXCOM, INC., San Diego, CA (US)
Filed by DexCom, Inc., San Diego, CA (US)
Filed on Jun. 19, 2020, as Appl. No. 16/906,812.
Claims priority of provisional application 62/863,648, filed on Jun. 19, 2019.
Prior Publication US 2020/0402633 A1, Dec. 24, 2020
Int. Cl. G16H 20/17 (2018.01); G06N 5/04 (2023.01); G16H 20/30 (2018.01); G16H 40/67 (2018.01); G16H 50/50 (2018.01); G16H 10/60 (2018.01); G16H 50/30 (2018.01); G16H 50/70 (2018.01); G16H 50/20 (2018.01); G16H 20/60 (2018.01)
CPC G16H 20/17 (2018.01) [G06N 5/04 (2013.01); G16H 10/60 (2018.01); G16H 20/30 (2018.01); G16H 20/60 (2018.01); G16H 40/67 (2018.01); G16H 50/20 (2018.01); G16H 50/30 (2018.01); G16H 50/50 (2018.01); G16H 50/70 (2018.01)] 25 Claims
OG exemplary drawing
 
1. A diabetes management system comprising:
at least one processor configured to perform calculations executing an input compiler, an exogenous pharmacon on board (XOB) estimator and an output compiler, the input compiler being configured to receive and process input data, wherein the XOB estimator estimates an amount of pharmacon that will produce a glucose response that is closest to a predicted glucose response due to historical pharmacon administrations;
wherein the XOB estimator comprises a baseline time-series estimator, an iterative time-series comparator, and a time-series comparator, wherein the baseline time-series estimator is a stateless machine that receives historical XOB amounts and continuous glucose monitoring (CGM) system history and produces a time-series that approximates future values of glucose, wherein the iterative time-series comparator is configured to combine a procedure that generates candidate XOB amounts with a modification of the time-series estimator in which historical XOB amounts are substituted with the XOB amount under consideration, wherein the time-series comparator is configured to match a pair of time-series that can be any measure of distance or similarity between the series; and
an insulin pump or pen injecting or infusing an amount of insulin to a patient based on the amount of pharmacon that is estimated for use in treatment of diabetes.