Optimization Of Thresholds For Eating Detection

• US 20090281449A1
• Filed: 10/23/2008
• Published: 11/12/2009
• Est. Priority Date: 05/09/2008
• Status: Active Grant

• Current Assignee: Tylerton International Inc.
• Original Assignee: MetaCure Limited (MetaCure (USA), Inc.)
• Inventors: Thrower, James, Goode, Paul, Policker, Shai, Kliger, Anat
• Granted Patent: US 8,423,130 B2
• Time in Patent Office: Days
• US Class Current: 600/547
• CPC Class Codes: A61B 5/0031 Implanted circuitry
• CPC Class Codes: A61B 5/053 Measuring electrical impedance or conductance of a portion of the body
• CPC Class Codes: A61B 5/4205 Evaluating swallowing

Abstract

A method is provided that includes providing a fundic impedance signal of a subject sensed during an initial calibration period using at least one implantable fundic sensor of an eating detection device. An indication is received of actual ingestion events of the subject during the initial calibration period. Using a cost function, an analysis is performed of the fundic impedance signal and the actual ingestion events. Responsively to the analysis, a threshold is set, the crossing of which by the fundic impedance signal is indicative of ingestion. The threshold is stored in a memory of the eating detection device. Other embodiments are also described.

Claims

1. A method comprising:

providing a fundic impedance signal of a subject sensed during an initial calibration period using at least one implantable fundic sensor of an eating detection device;

receiving an indication of actual ingestion events of the subject during the initial calibration period;

using a cost function, performing an analysis of the fundic impedance signal and the actual ingestion events;

responsively to the analysis, setting a threshold, the crossing of which by the fundic impedance signal is indicative of ingestion; and

storing the threshold in a memory of the eating detection device.

2. The method according to claim 1,

wherein providing the fundic impedance signal comprises:

receiving a sensed fundic impedance signal of the subject sensed during the initial calibration period using the at least one implantable fundic sensor; and

filtering the sensed fundic impedance signal to generate a filtered fundic impedance signal, and

wherein performing the analysis comprises performing the analysis of the filtered fundic impedance signal and the actual ingestion events.

3. The method according to claim 2, wherein receiving the sensed fundic impedance signal comprises receiving a representative sample of a raw fundic impedance signal, and wherein filtering comprises filtering the representative sample of the raw fundic impedance signal to generate the filtered fundic impedance signal.

4. The method according to claim 1, wherein providing the fundic impedance signal comprises receiving a filtered fundic impedance signal from the eating detection device, and wherein performing the analysis comprises performing the analysis of the filtered fundic impedance signal and the actual ingestion events.

5. The method according to claim 1, wherein performing the analysis and setting the threshold comprise performing the analysis and setting the threshold by the eating detection device.

6. The method according to claim 1, wherein performing the analysis and setting the threshold comprise performing the analysis and setting the threshold by a computer workstation separate from the eating detection device, and wherein storing the threshold in the memory of the eating detection device comprises transmitting, by the workstation, the threshold to the eating detection device.

7. The method according to claim 1,

wherein the threshold is one of a plurality of candidate thresholds,

wherein performing the analysis comprises using the cost function to find each of a plurality of cost function values for each of the respective candidate thresholds by:

identifying detected ingestion events during the initial calibration period by comparing the fundic impedance signal to the candidate threshold over the initial calibration period;

comparing the detected ingestion events to the actual ingestion events to find a number of the detected ingestion events that represent true positives, a number of the detected ingestion events that represent false positives, and a number of the actual ingestion events that represent false negatives; and

calculating the cost function value responsively to the number of the true positives, the number of the false positives, and the number of the false negatives, and

wherein setting the threshold comprises selecting one of the candidate thresholds as the threshold responsively to the cost function values.

8. The method according to claim 7, wherein selecting the one of the candidate thresholds comprises selecting the one of the candidate thresholds responsively to a ranking of the cost function values.

11. The method according to claim 7, wherein calculating the cost function value comprises calculating the cost function value responsively to a measure of sensitivity of the algorithm and a measure of positive predictive value of the algorithm.

12. The method according to claim 11, wherein calculating the cost function value comprises calculating the measure of sensitivity responsively to the number of the true positives and the number of the false negatives.

13. The method according to claim 12, wherein calculating the measure of sensitivity comprises calculating the quotient of the number of the true positives divided by a sum of the number of the true positives and the number of the false negatives.

14. The method according to claim 11, wherein calculating the cost function value comprises calculating the measure of positive predictive value responsively to the number of the true positives and the number of the false positives.

15. The method according to claim 14, wherein calculating the measure of positive predictive value comprises calculating the quotient of the number of the true positives divided by a sum of the number of the true positives and the number of the false positives.

16. The method according to claim 11, wherein calculating the cost function comprises calculating a sum of the measure of sensitivity and a product of the measure of positive predictive value and a weighting factor.

17. The method according to claim 16, wherein the weighting factor is between 0.7 and 1.3.

21. The method according to claim 1, further comprising, during a treatment period after the initial calibration period:

sensing, using the at least one fundic sensor of the eating detection device, a treatment sensed fundic impedance signal;

filtering, by the eating detection device, the treatment sensed fundic impedance signal to generate a treatment filtered fundic impedance signal; and

identifying, by the eating detection device, a treatment period ingestion event by comparing the treatment filtered fundic impedance signal to the selected one of the thresholds stored in the memory of the eating detection device.

22. The method according to claim 21, further comprising, responsively to identifying the treatment period ingestion event, applying a treatment to the subject by the eating detection device.

23. The method according to claim 1, wherein the actual ingestion events are actual solid food ingestion events, and wherein receiving the indication of the actual ingestion events comprises receiving the indication of the actual solid food ingestion events by the subject during the initial calibration period.

24. The method according to claim 1, wherein providing the fundic impedance signal comprises providing a set of filtered fundic impedance increase vs. baseline values, and wherein setting the threshold comprises setting a fundic impedance increase vs. baseline threshold.

9-10. -10. (canceled)

18-20. -20. (canceled)

25. Apparatus for use with an eating detection device having a memory and at least one implantable fundic sensor, the apparatus comprising:

an interface, configured to communicate with the eating detecting device; and

a processor, configured to receive, via the interface, a fundic impedance signal of a subject sensed during an initial calibration period using the at least one implantable fundic sensor of the eating detection device;

to receive an indication of actual ingestion events of the subject during the initial calibration period;

using a cost function, to perform an analysis of the fundic impedance signal and the actual ingestion events;

responsively to the analysis, to set a threshold, the crossing of which by the fundic impedance signal is indicative of ingestion; and

to store, via the interface, the threshold in the memory of the eating detection device.

26. The apparatus according to claim 25, wherein the processor is configured to filter the fundic impedance signal to generate a filtered fundic impedance signal, and to perform the analysis of the filtered fundic impedance signal and the actual ingestion events.

27. The apparatus according to claim 25, wherein the fundic impedance signal includes a filtered fundic impedance signal, and wherein the processor is configured to receive the filtered fundic impedance signal from the eating detection device, via the interface, and to perform the analysis of the filtered fundic impedance signal and the actual ingestion events.

28. The apparatus according to claim 25,

wherein the threshold is one of a plurality of candidate thresholds,

wherein the processor is configured to perform the analysis by using the cost function to find each of a plurality of cost function values for each of the respective candidate thresholds by identifying detected ingestion events during the initial calibration period by comparing the fundic impedance signal to the candidate threshold over the initial calibration period;

comparing the detected ingestion events to the actual ingestion events to find a number of the detected ingestion events that represent true positives, a number of the detected ingestion events that represent false positives, and a number of the actual ingestion events that represent false negatives; and

calculating the cost function value responsively to the number of the true positives, the number of the false positives, and the number of the false negatives, and

wherein the processor is configured to set the threshold by selecting one of the candidate thresholds as the threshold responsively to the cost function values.

29. The apparatus according to claim 28, wherein the processor is configured to select the one of the candidate thresholds responsively to a ranking of the cost function values.

30. The apparatus according to claim 28, wherein the processor is configured to calculate the cost function value responsively to a measure of sensitivity of the algorithm and a measure of positive predictive value of the algorithm.

31. The apparatus according to claim 30, wherein the processor is configured to calculate the measure of sensitivity responsively to the number of the true positives and the number of the false negatives.

32. The apparatus according to claim 31, wherein the processor is configured to calculate the measure of sensitivity by calculating the quotient of the number of the true positives divided by a sum of the number of the true positives and the number of the false negatives.

33. The apparatus according to claim 30, wherein the processor is configured to calculate the measure of positive predictive value responsively to the number of the true positives and the number of the false positives.

34. The apparatus according to claim 33, wherein the processor is configured to calculate the measure of positive predictive value by calculating the quotient of the number of the true positives divided by a sum of the number of the true positives and the number of the false positives.

35. The apparatus according to claim 30, wherein the processor is configured to calculate the cost function by calculating a sum of the measure of sensitivity and a product of the measure of positive predictive value and a weighting factor.

38. The apparatus according to claim 25, wherein the actual ingestion events are actual solid food ingestion events, and wherein the processor is configured to receive the indication of the actual solid food ingestion events by the subject during the initial calibration period.

36-37. -37. (canceled)

39. Apparatus comprising:

one or more implantable fundic sensors, configured to be applied to a fundus of a subject, and to generate a fundic impedance signal; and

a control element, which comprises a memory, and which is configured to;

receive the fundic impedance signal sensed during an initial calibration period,

receive an indication of actual ingestion events of the subject during the initial calibration period,

using a cost function, perform an analysis of the fundic impedance signal and the actual ingestion events,

responsively to the analysis, set a threshold, the crossing of which by the fundic impedance signal is indicative of ingestion, and store the threshold in the memory.

40. The apparatus according to claim 39, wherein the control element is configured to filter the fundic impedance signal to generate a filtered fundic impedance signal, and perform the analysis of the filtered fundic impedance signal and the actual ingestion events.

41. The apparatus according to claim 39,

wherein the threshold is one of a plurality of candidate thresholds,

wherein the control element is configured to perform the analysis by using the cost function to find each of a plurality of cost function values for each of the respective candidate thresholds by identifying detected ingestion events during the initial calibration period by comparing the fundic impedance signal to the candidate threshold over the initial calibration period;

comparing the detected ingestion events to the actual ingestion events to find a number of the detected ingestion events that represent true positives, a number of the detected ingestion events that represent false positives, and a number of the actual ingestion events that represent false negatives; and

calculating the cost function value responsively to the number of the true positives, the number of the false positives, and the number of the false negatives, and

wherein the control element is configured to set the threshold by selecting one of the candidate thresholds as the threshold responsively to the cost function values.

42. The apparatus according to claim 41, wherein the control element is configured to selecting the one of the candidate thresholds responsively to a ranking of the cost function values.

43. The apparatus according to claim 41, wherein the control element is configured to calculate the cost function value responsively to a measure of sensitivity of the algorithm and a measure of positive predictive value of the algorithm.

46. The apparatus according to claim 43, wherein the control element is configured to calculate the cost function value by calculating the measure of positive predictive value responsively to the number of the true positives and the number of the false positives.

47. The apparatus according to claim 46, wherein the control element is configured to calculate the measure of positive predictive value by calculating the quotient of the number of the true positives divided by a sum of the number of the true positives and the number of the false positives.

48. The apparatus according to claim 43, wherein the control element is configured to calculate the cost function by calculating a sum of the measure of sensitivity and a product of the measure of positive predictive value and a weighting factor.

51. The apparatus according to claim 39, wherein the control element is configured to, during a treatment period after the initial calibration period, receive, from the at least one fundic sensor, a treatment sensed fundic impedance signal, filter the treatment sensed fundic impedance signal to generate a treatment filtered fundic impedance signal, and identify a treatment period ingestion event by comparing the treatment filtered fundic impedance signal to the selected one of the thresholds stored in the memory of the eating detection device.

52. The apparatus according to claim 51, further comprising an implantable treatment unit, wherein the control element is configured to drive the implantable treatment unit to apply a treatment to the subject, responsively to identifying the treatment period ingestion event.

53. The apparatus according to claim 39, wherein the actual ingestion events are actual solid food ingestion events, and wherein the control element is configured to receive the indication of the actual solid food ingestion events by the subject during the initial calibration period.

44-45. -45. (canceled)

49-50. -50. (canceled)

54. A computer software product for use with an eating detection device having a memory and at least one implantable fundic sensor, the product comprising a tangible computer-readable medium in which program instructions are stored, which instructions, when read by a computer, cause the computer to receive a fundic impedance signal of a subject sensed during an initial calibration period using the at least one implantable fundic sensor of the eating detection device;

to receive an indication of actual ingestion events of the subject during the initial calibration period;

using a cost function, to perform an analysis of the fundic impedance signal and the actual ingestion events;

responsively to the analysis, to set a threshold, the crossing of which by the fundic impedance signal is indicative of ingestion; and

to store the threshold in the memory of the eating detection device.

55. The computer software product according to claim 54, wherein the threshold is one of a plurality of candidate thresholds, wherein the instructions, when read by the computer, cause the computer to perform the analysis by using the cost function to find each of a plurality of cost function values for each of the respective candidate thresholds by identifying detected ingestion events during the initial calibration period by comparing the fundic impedance signal to the candidate threshold over the initial calibration period;

comparing the detected ingestion events to the actual ingestion events to find a number of the detected ingestion events that represent true positives, a number of the detected ingestion events that represent false positives, and a number of the actual ingestion events that represent false negatives; and

calculating the cost function value responsively to the number of the true positives, the number of the false positives, and the number of the false negatives, and wherein the instructions, when read by the computer, cause the computer to set the threshold by selecting one of the candidate thresholds as the threshold responsively to the cost function values.

56. The computer software product according to claim 55, wherein the instructions, when read by the computer, cause the computer to select the one of the candidate thresholds responsively to a ranking of the cost function values.

57. The computer software product according to claim 55, wherein the instructions, when read by the computer, cause the computer to calculate the cost function value responsively to a measure of sensitivity of the algorithm and a measure of positive predictive value of the algorithm.