System and method for detecting decoupling of an auscultatory sound sensor from a test-subject

1. A method, comprising:

a. acquiring at least one block of time-series data, wherein each said at least one block of time-series data comprises a plurality of samples of an auscultatory-sound signal generated by an auscultatory-sound sensor operatively coupled to a portion of the skin of a test-subject;

b. determining at least one comparison value selected from the group consisting of at least one data-dependent scale factor and at least one data-dependent detection threshold, wherein the operation of determining said at least one comparison value is responsive to said at least one block of time-series data, and said at least one comparison value provides a measure of a range of values of said at least one block of time-series data in relation to a predetermined metric; and

c. determining whether or not said auscultatory-sound sensor is either debonded or detached from said skin of said test-subject responsive to at least one of said at least one comparison value.

2. A method as recited in claim 1 , wherein the operation of determining at least one of said at least one comparison value comprises determining said at least one data-dependent scale factor responsive to a corresponding said at least one block of time-series data, wherein the value of said at least one data-dependent scale factor is determined so that the values of said plurality of samples of said corresponding at least one block of time-series data, when multiplied by said at least one data-dependent scale factor, are nominally within a dynamic range of an associated data acquisition system used to acquire said at least one block of time-series data.

3. A method as recited in claim 2 , wherein the operation of determining said at least one data-dependent scale factor comprises:

a. segmenting said at least one block of time-series data into a plurality of time-series segments;

b. determining a plurality of peak absolute values of said auscultatory-sound signal, wherein each peak absolute value of said plurality of peak absolute values of said auscultatory-sound signal corresponds to a peak absolute value of said auscultatory-sound signal within a corresponding time-series segment of said plurality of time-series segments; and

c. determining a median value of said plurality of peak absolute values of said auscultatory-sound signal, wherein said at least one data-dependent scale factor is inversely related to said median value of said plurality of peak absolute values of said auscultatory-sound signal.

4. A method as recited in claim 3 , wherein a time period spanned by each time-series segment of said plurality of time-series segments of said at least one block of time-series data is nominally about one heart cycle.

5. A method as recited in claim 3 , wherein said at least one data-dependent scale factor is responsive to a ratio of said dynamic range of said associated data acquisition system used to acquire said auscultatory-sound signal, divided by said median value of said plurality of peak absolute values of said auscultatory-sound signal.

6. A method as recited in claim 5 , wherein said dynamic range is a fraction of a zero-to-peak range of said associated data acquisition system.

7. A method as recited in claim 6 , wherein said fraction is in the range of 0.65 to 0.95.

8. A method as recited in claim 2 , further comprising comparing said at least one data-dependent scale factor with a detachment threshold, and determining whether or not said auscultatory-sound sensor is detached from said skin of said test-subject responsive to whether or not said at least one data-dependent scale factor exceeds said detachment threshold.

9. A method as recited in claim 8 , wherein if said auscultatory-sound sensor is determined to be detached from said skin of said test-subject, then either aborting the acquisition, or nullifying a recording, of said at least one block of time-series data.

10. A method as recited in claim 2 , wherein the operation of determining at least one said at least one data-dependent scale factor responsive to said corresponding said at least one block of time-series data comprises determining a nominal data-dependent scale factor responsive to one block of time-series data of said at least one block of time-series data, and the operation of determining whether or not said auscultatory-sound sensor is either debonded or detached from said skin of said test-subject is both responsive to said nominal data-dependent scale factor determined from said one block of time-series data, and responsive to one or more blocks of said time-series data recorded while said test-subject was holding their breath.

11. A method as recited in claim 10 , wherein the operation of determining whether or not said auscultatory-sound sensor is debonded from said skin of said test-subject comprises, for each successive value of successive samples of said auscultatory sound signal, from said one or more blocks of said time-series data:

a. for each of said successive samples of said auscultatory sound signal, comparing a scaled value of said successive value with a predetermined detection threshold, wherein said scaled value is determined by multiplying said successive value by said nominal data-dependent scale factor; and

d. providing an indication that said auscultatory-sound sensor is debonded from said skin of said test-subject if the number of said successive samples of said auscultatory sound signal for which said scaled value is less than said predetermined detection threshold is greater than a count threshold.

12. A method as recited in claim 11 , wherein the operation of comparing said scaled value with said predetermined detection threshold comprises comparing an absolute value of said scaled value with said predetermined detection threshold.

13. A method as recited in claim 11 , wherein the operation of comparing said scaled value with said predetermined detection threshold is performed for each sample during the operation of acquiring said at least one block of time-series data.

14. A method as recited in claim 11 , wherein if said auscultatory-sound sensor is determined to be debonded from said skin of said test-subject, then either aborting the acquisition, or nullifying a recording, of said at least one block of time-series data.

15. A method as recited in claim 2 , wherein the operation of determining at least one said at least one data-dependent scale factor responsive to said corresponding said at least one block of time-series data comprises determining a plurality of data-dependent scale factors, wherein each data-dependent scale factor of said plurality of data-dependent scale factors is responsive to a corresponding associated said at least one block of time-series data in one-to-one relationship therewith, and the operation of determining whether or not said auscultatory-sound sensor is either debonded or detached from said skin of said test-subject is both responsive to at least one said data-dependent scale factor, and responsive to said corresponding associated said at least one said block of time-series data recorded while said test-subject was holding their breath.

16. A method as recited in claim 1 , wherein the operation of determining at least one of said at least one comparison value comprises determining said at least one data-dependent detection threshold responsive to a corresponding said at least one block of time-series data, wherein the value of said at least one data-dependent detection threshold is responsive to a nominal detection threshold scaled by a measure responsive to a range of values of corresponding samples said at least one block of time-series data, wherein said nominal detection threshold is responsive to a dynamic range of an associated data acquisition system used to acquire said at least one block of time-series data.

17. A method as recited in claim 16 , wherein the operation of determining said at least one data-dependent detection threshold comprises:

a. segmenting said at least one block of time-series data into a plurality of time-series segments;

b. determining a plurality of peak absolute values of said auscultatory-sound signal, wherein each peak absolute value of said plurality of peak absolute values of said auscultatory-sound signal corresponds to a peak absolute value of said auscultatory-sound signal within a corresponding time-series segment of said plurality of time-series segments;

c. determining a median value of said plurality of peak absolute values of said auscultatory-sound signal, wherein said at least one data-dependent detection threshold is directly related to said median value of said plurality of peak absolute values of said auscultatory-sound signal.

18. A method as recited in claim 17 , wherein a time period spanned by each time-series segment of said plurality of time-series segments of said at least one block of time-series data is nominally about one heart cycle.

19. A method as recited in claim 17 , wherein said at least one data-dependent detection threshold is inversely responsive to a ratio of said dynamic range of said associated data acquisition system used to acquire said auscultatory-sound signal, divided by said median value of said plurality of peak absolute values of said auscultatory-sound signal.

20. A method as recited in claim 19 , wherein said dynamic range is a fraction of a zero-to-peak range of said associated data acquisition system.

21. A method as recited in claim 20 , wherein said fraction is in the range of 0.65 to 0.95.

22. A method as recited in claim 16 , further comprising comparing said at least one data-dependent detection threshold with a detachment threshold, and determining whether or not said auscultatory-sound sensor is detached from said skin of said test-subject responsive to whether or not said at least one data-dependent detection threshold is less then said detachment threshold.

23. A method as recited in claim 22 , wherein if said auscultatory-sound sensor is determined to be detached from said skin of said test-subject, then either aborting the acquisition, or nullifying a recording, of said at least one block of time-series data.

24. A method as recited in claim 16 , wherein the operation of determining whether or not said auscultatory-sound sensor is either debonded or detached from said skin of said test-subject is both responsive to at least one said at least one data-dependent detection threshold, and responsive to one or more blocks of said time-series data recorded while said test-subject was holding their breath.

25. A method as recited in claim 16 , wherein the operation of determining whether or not said auscultatory-sound sensor is debonded from said skin of said test-subject comprises, for each successive value of successive samples of said auscultatory sound signal, from one or more blocks of said time-series data:

d. for each of said successive samples of said auscultatory sound signal, comparing said successive value with said at least one data-dependent detection threshold, and

e. providing an indication that said auscultatory-sound sensor is debonded from said skin of said test-subject if the number of said successive samples of said auscultatory sound signal for which said successive value is less than said at least one data-dependent detection threshold is greater than a count threshold.

26. A method as recited in claim 25 , wherein the operation of comparing said successive value with said at least one data-dependent detection threshold comprises comparing an absolute value of said successive value with said at least one data-dependent detection threshold.

27. A method as recited in claim 25 , wherein if said auscultatory-sound sensor is determined to be debonded from said skin of said test-subject, then either aborting the acquisition, or nullifying a recording, of said at least one block of time-series data.