Pattern recognition and filtering in a therapy management system

• US 9,330,237 B2
• Filed: 12/24/2008
• Issued: 05/03/2016
• Est. Priority Date: 12/24/2008
• Status: Active Grant

• Current Assignee: Medtronic Minimed Incorporated (Medtronic Plc)
• Original Assignee: Medtronic Minimed Incorporated (Medtronic Plc)
• Inventors: Cohen, Gary, Getschmann, Kristen, Raman, Vidya, Ko, Kenneth, Nair, Biju, Talbot, Cary, Nitzan, Tamir, Kovelman, Paul H., Hess, Bee
• Primary Examiner(s): Thomson, Bill
• Assistant Examiner(s): Sands, Davin K
• Application Number:
  US12/343,875
• Publication Number:
  US 20100161236A1
• Time in Patent Office: 2,687 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: A61B 5/4836 Diagnosis combined with treatment in closed-loop systems or methods A61B5/0036 takes precedence
• CPC Class Codes: A61B 5/7275 Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: A61B 5/7282 Event detection, e.g. detecting unique waveforms indicative of a medical condition A61B5/4094, A61B5/4818, A61B5/349, A61B5/0823 take precedence
• CPC Class Codes: A61M 5/1723 using feedback of body parameters, e.g. blood-sugar, pressure measurement of body parameters A61B5/00
• CPC Class Codes: G16H 20/10 relating to drugs or medications, e.g. for ensuring correct administration to patients
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

A method of diabetes analysis includes receiving a plurality of glucose level readings for a user. A common event occurrence in at least two of the glucose level readings is determined. The at least two glucose level readings from the common event occurrence onwards in time for a time period is analyzed. A glucose level pattern formed by the at least two glucose level readings having a similar shape is determined. At least one anomalous glucose level reading having the similar shape and not conforming to the glucose level pattern is analyzed. The at least one anomalous glucose level reading is adapted to the pattern to form an adapted glucose level pattern. An insulin dosage for the time period beginning at the common event occurrence is calculated based on the adapted glucose level pattern.

Claims

1. A method of diabetes analysis, comprising:

receiving, using a processor and a glucose sensor, a plurality of glucose level readings over a plurality of days for a user, each of the plurality of glucose level readings having multiple glucose level measurements for one of the plurality of days;

determining, using the processor, a common event occurrence in at least two of the glucose level readings;

analyzing, using the processor, the at least two glucose level readings from the common event occurrence onwards in time for a time period;

identifying, using the processor, a glucose level pattern formed by the at least two glucose level readings having a similar shape;

analyzing, using the processor, an anomalous glucose level reading not conforming to the glucose level pattern to determine whether the anomalous glucose level reading has the similar shape;

adjusting, using the processor, the anomalous glucose level reading to adapt to the glucose level pattern to form an adapted glucose level pattern if the anomalous glucose level reading has the similar shape;

filtering out, using the processor, the anomalous glucose level reading if the anomalous glucose level reading does not have the similar shape;

calculating, using the processor, an insulin dosage for the time period beginning at the common event occurrence based on the adapted glucose level pattern; and

initiating an action, wherein the action includes at least one of recommending the calculated dosage and automatically delivering the calculated insulin dosage,

wherein the method is implemented on a portable medical system.

2. The method of claim 1, wherein the common event occurrence is selected from the group consisting of:

breakfast, lunch, dinner, a meal bolus, a correction bolus, and a bedtime.

3. The method of claim 1, wherein the insulin dosage is for a basal insulin dosage.

4. The method of claim 1, wherein the at least one anomalous glucose level reading having the similar shape has at least one of:

a greater or lesser magnitude, and a higher or lower basal glucose level than the at least two glucose level readings forming the glucose level pattern.

5. The method of claim 1, wherein adjusting, using the processor, the anomalous glucose level reading to adapt to the glucose level pattern if the anomalous glucose level reading has the similar shape includes at least one of:

shifting the anomalous glucose level reading in time compared to the at least two glucose level readings forming the glucose level pattern, normalizing a magnitude of the anomalous glucose level reading toward a target range of glucose level measurements, compressing or stretching the anomalous glucose level reading in time compared to the at least two glucose level readings forming the glucose level pattern, and shifting the anomalous glucose level reading toward the target range of glucose level measurements.

6. The method of claim 1, wherein the plurality of glucose level readings for the user excludes readings from at least one recent day.

7. The method of claim 1, wherein the plurality of glucose level readings are selected only from days that the user has a periodic or transient condition.

8. An article of manufacture containing code for diabetes analysis, comprising a computer-usable medium including at least one embedded computer program that is capable of causing at least one computer to perform:

receiving a plurality of glucose level readings over a plurality of days for a user from a glucose sensor, each of the plurality of glucose level readings having multiple glucose level measurements for one of the plurality of days;

determining a common event occurrence in at least two of the glucose level readings;

analyzing the at least two glucose level readings from the common event occurrence onwards in time for a time period;

identifying a glucose level pattern formed by the at least two glucose level readings having a similar shape;

analyzing an anomalous glucose level reading not conforming to the glucose level pattern to determine whether the anomalous glucose level reading has the similar shape;

adjusting the anomalous glucose level reading to adapt to the glucose level pattern to form an adapted glucose level pattern if the anomalous glucose level reading has the similar shape;

filtering out the anomalous glucose level reading if the anomalous glucose level reading does not have the similar shape;

calculating an insulin dosage for the period beginning at the common event occurrence based on the adapted glucose level pattern; and

initiating an action, wherein the action includes at least one of recommending the calculated insulin dosage and automatically delivering the calculated insulin dosage,

wherein the article of manufacture is a portable medical system.

9. The article of claim 8, wherein the common event occurrence is selected from the group consisting of:

breakfast, lunch, dinner, a meal bolus, a correction bolus, and a bedtime.

10. The article of claim 8, wherein the insulin dosage is for a basal insulin dosage.

11. The article of claim 8, wherein the anomalous glucose level reading having the similar shape has at least one of:

a greater or lesser magnitude, and a higher or lower basal glucose level than the at least two glucose level readings forming the glucose level pattern.

12. The article of claim 8, wherein adjusting the anomalous glucose level reading to adapt to the glucose level pattern if the anomalous glucose reading has the similar shape includes at least one of:

shifting the anomalous glucose level reading in time compared to the at least two glucose level readings forming the glucose level pattern, normalizing a magnitude of the anomalous glucose level reading toward a target range of glucose level measurements, compressing or stretching the anomalous glucose level reading in time compared to the at least two glucose level readings forming the glucose level pattern, and shifting the anomalous glucose level reading toward the target range of glucose level measurements.

13. The article of claim 8, wherein the plurality of glucose level readings for the user excludes readings from at least one recent day.

14. The article of claim 8, wherein the plurality of glucose level readings are selected only from days that the user has a periodic or transient condition.