Systems and methods for detecting glucose level data patterns

Patent metadata

Field	Value
Publication number	US-10349871-B2
Application number	US-201213566678-A
Country	US
Kind code	B2
Filing date	Aug 3, 2012
Priority date	Aug 5, 2011
Publication date	Jul 16, 2019
Grant date	Jul 16, 2019

Abstract

Systems and methods for detecting and reporting patterns in analyte concentration data are provided. According to some implementations, an implantable device for continuous measurement of an analyte concentration is disclosed. The implantable device includes a sensor configured to generate a signal indicative of a concentration of an analyte in a host, a memory configured to store data corresponding at least one of the generated signal and user information, a processor configured to receive data from at least one of the memory and the sensor, wherein the processor is configured to generate pattern data based on the received information, and an output module configured to output the generated pattern data. The pattern data can be based on detecting frequency and severity of analyte data in clinically risky ranges.

First claim

What is claimed is:

1. A computer-implemented method for identifying patterns in continuous analyte data of a patient user monitored by a continuous analyte monitoring system comprising a continuous analyte sensor device to generate the continuous analyte data in communication with a computing device including a memory to store the generated continuous analyte data and a processor module to process the continuous analyte data according to the computer-implemented method, the computer-implemented method comprising:
obtaining analyte data points falling within a designated date range, wherein an analyte data point includes an analyte value and a corresponding time value;
applying one of more filters to the analyte data points to generate contributor data points;
weighting each contributor data point of the generated contributor data points based on the contributor data point's analyte value;
assigning each weighted contributor data point to a corresponding epoch according to the time value associated with the weighted contributor data point;
applying at least one pattern threshold to an epoch to determine a match of whether the weighted contributor data point or weighted contributor data points assigned to the epoch meet the at least one pattern threshold, wherein the applying includes assigning a flag to the epoch when a match is determined;
determining one or more pattern occurrences by scanning the epochs for flags, wherein the determined one or more pattern occurrences are indicative of repeated patterns of the patient user's monitored analyte levels; and
outputting information representative of the determined one or more pattern occurrences, wherein the outputted pattern information is transmitted to an insulin pump controller based on an automatic trigger to control insulin administration based on the pattern information.

2. The method of claim 1 , wherein the designated date range is at least three days.

3. The method of claim 1 , wherein the one or more filters comprises filters selected from the group consisting of:
(i) an analyte value filter that filters out analyte data points that have an analyte concentration value falling outside of a predetermined analyte level or range of analyte values;
(ii) a time range filter that filters out analyte data points measurements that falls outside of a time of day range;
(iii) a day of the week filter that filters out analyte data points measured one or more predetermined days of the week; and
(iv) an event filter that filters out data points that do not fall within a predetermined amount of time from an occurrence of an event.

4. The method of claim 3 , wherein the event is an event selected from the group consisting of exercise, a meal, sleep and administration of a medication.

5. The method of claim 4 , further comprising receiving user input indicative of the event using a user interface.

6. The method of claim 1 , wherein weighting each contributor data point is based on a predetermined weight assignment map or mathematical assignment function.

7. The method of claim 1 , wherein weighting each contributor data point comprises assigning a weighted value to each contributor data point, wherein the assigned weighted value is smaller if the analyte value is less clinically significant than if the analyte value is more clinically significant.

8. The method of claim 7 , wherein the assignment of weighted values is non-linear based on the analyte value.

9. The method of claim 1 , wherein each epoch spans a determined time of day and wherein each contributor data point is added to the epoch that spans the corresponding time of day.

10. The method of claim 1 , wherein the pattern thresholds comprise thresholds selected from the group consisting of:
a threshold minimum number of contributor data points in an epoch;
a threshold average weighted value of the contributor data points in an epoch;
a threshold medium weighted value of the contributor data points in the epoch;
a threshold sum of the weighted values of the contributor data points in the epoch;
a threshold average difference of the weighted values of the contributor data points in the epoch;
a threshold standard deviation value of the weighted values of the contributor data points in the epoch; and
a threshold correlation value.

11. The method of claim 10 , wherein at least one of the thresholds is defined in terms of a percentage.

12. The method of claim 1 , wherein an epoch is flagged when at least some of the one or more thresholds are satisfied.

13. The method of claim 1 , wherein determining the one or more pattern occurrences comprises determining a start and an end of a pattern.

14. The method of claim 13 , wherein determining the start and the end of the pattern is determined based on identifying a predetermined minimum number of contiguous matching epochs or a predetermined ratio of contiguous matching to non-matching epochs.

15. The method of claim 13 , wherein the start of the pattern is determined based on identifying a first threshold number of contiguous matching epochs and the end of the pattern is determined based on identifying a second threshold number non-matching epochs.

16. The method of claim 13 , wherein the outputted information comprises the start time and the end time associated with when a repeated pattern of the patient user's monitored analyte levels occurred.

17. The method of claim 1 , wherein determining the one or more pattern occurrences is based on the frequency of matching epochs and the weighted values of the matching epochs.

18. The method of claim 1 , wherein the outputted information comprises a triggering alert notifying the patient user of the one or more patterns occurrences.

19. The method of claim 1 , wherein the outputted information is processed further to form or modify a medication administration routine.

20. A method for augmenting the performance of a continuous analyte monitoring system by identifying repeated patterns in continuous analyte data of a patient user monitored by the continuous analyte monitoring system, in which the continuous analyte monitoring system comprises a continuous analyte sensor device to generate the continuous analyte data in communication with a computing device including a memory to store the generated continuous analyte data and a processor module to process the continuous analyte data to identify the repeated patterns, the method comprising:
(a) detecting, by the continuous analyte sensor device, analyte levels of the patient user over a designated time range to form a continuous analyte data set;
(b) processing, by the computing device, the continuous analyte data set to identify repeated patterns in the detected analyte values over the designated time range, the processing including:
(i) obtaining analyte data points of the continuous analyte data set falling within a designated time range, wherein an analyte data point includes an analyte value and a corresponding time value,
(ii) applying one of more filters to the analyte data points to generate contributor data points,
(iii) weighting each contributor data point of the generated contributor data points based on the contributor data point's analyte value,
(iv) assigning each weighted contributor data point to a corresponding epoch according to the time value associated with the weighted contributor data point,
(v) applying at least one pattern threshold to an epoch to determine a match of whether the weighted contributor data point or weighted contributor da

Assignees

• Mayou Phil
• Hampapuram Hari
• Price David
• Weindel Keri
• Snisarenko Kostyantyn
• Mensinger Michael Robert
• Bowman Leif N
• Boock Robert J
• Kamath Apurv Ullas
• Reihman Eli
• Simpson Peter C
• Dexcom Inc

Inventors

• Mayou Phil
• Hampapuram Hari
• Price David
• Weindel Keri
• Snisarenko Kostyantyn
• Mensinger Michael Robert
• Bowman Leif N
• Boock Robert J
• Kamath Apurv Ullas
• Reihman Eli
• Simpson Peter C

Classifications

• A61B5/14532 Primary

  for measuring glucose, e.g. by tissue impedance measurement · CPC title

• G16H50/70

  for mining of medical data, e.g. analysing previous cases of other patients · CPC title

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• G16H15/00

  ICT specially adapted for medical reports, e.g. generation or transmission thereof · CPC title

• Y02A90/26

  Cross-Sectional Technologies · mapped topic