Method, system and computer program method for detection of pathological fluctuations of physiological signals to diagnose human illness

• US 8,615,291 B2
• Filed: 03/15/2010
• Issued: 12/24/2013
• Est. Priority Date: 03/13/2009
• Status: Active Grant

• Current Assignee: National Institutes of Health (NIH) (Government of the United States of America), United States Department Of Health And Human Services (Government of the United States of America)
• Original Assignee: National Institutes of Health (NIH) (Government of the United States of America), United States Department Of Health And Human Services (Government of the United States of America)
• Inventors: Lake, Douglas E., Flower, Abigail, Delos, John B., Moorman, Randall
• Primary Examiner(s): PATTON, AMANDA K
• Application Number:
  US12/724,162
• Publication Number:
  US 20100234748A1
• Time in Patent Office: 1,380 Days
• Field of Search: 600515-518, 600/521, 600/509
• US Class Current: 600/509
• CPC Class Codes: A61B 5/02405 Determining heart rate variability
• CPC Class Codes: A61B 5/35 by template matching
• CPC Class Codes: A61B 5/412 Detecting or monitoring sepsis
• 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: G06N 7/02 using fuzzy logic computing arrangements based on biological models G06N3/00; computing arrangements using knowledge-based models G06N5/00
• CPC Class Codes: G16H 10/20 for electronic clinical trials or questionnaires
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders

Abstract

Method, system, and computer program method for detecting pathological fluctuations of physiological signals to diagnose human illness. The method comprises performing a sliding window analysis to find sequences in physiological signal data that match amplitude- and duration-adjusted versions of a template function to within a specified tolerance.

Claims

1. A method comprising:

accepting time series data of a cardiac signal including a heart beat or an RR interval;

obtaining a pattern for the time series data;

choosing a template function that corresponds to the pattern of the time-series data;

performing an analysis of the time series data to match sequences in the time series data to the template function, wherein one or more of the sequences comprises fluctuations;

calculating one or more characteristics of the fluctuations based on the analysis;

identifying a risk of clinical condition associated with one or more of the characteristics.

2. The method of claim 1, wherein the fluctuations include accelerations or decelerations.

3. The method of claim 1, wherein the template function is a wavelet function.

4. The method of claim 3, wherein the wavelet function is selected from a group consisting of exponential, Gaussian, Lorentzian, and χ

(n), wherein χ

(n) is

5. The method of claim 4, wherein the template function is χ

(n).

6. The method of claim 1, wherein the analysis comprises a sliding-window analysis, the analysis including sweeping the template function through the time series data in sequence increments and calculating a correlation coefficient between the template function and any sequence increment, wherein the sequence increment may comprise one or more fluctuations.

7. The method of claim 6, wherein different analytic forms of the template function are swept through the time series data.

8. The method of claim 7, wherein the different analytic forms of the template function that are swept through the time series data vary in amplitude, duration, or a combination thereof.

9. The method of claim 8, wherein the different analytic form of the template function that is swept through the time series data increase in duration from about 8 to about 100 beats.

10. The method of claim 7, further comprising determining the different analytic forms of the template function that highly correlate with the sequence increments.

11. The method of claim 10, wherein the correlation between the different analytic forms of the template function and the sequence increments is at a minimum threshold.

12. The method of claim 10, further comprising removing overlapping fluctuations.

13. The method of claim 12, wherein removal of overlapping fluctuations comprises identifying and keeping the overlapping fluctuation that has the highest correlation coefficient.

14. The method of claim 1, further comprising removing the fluctuations from the cardiac signal and determining baseline signal variability.

15. The method of claim 14, wherein baseline signal variability is measured through sample asymmetry, smoothness, residual variability, or a combination thereof.

16. The method of claim 1, wherein the characteristics of the fluctuations are selected from the group consisting of number of fluctuations;

amplitude of the fluctuations;

widths of the fluctuations;

R² between the template function and fluctuation; and

baseline signal variability.

17. The method of claim 1, wherein the clinical condition is an illness.

18. The method of claim 17, wherein the illness is sepsis or a condition associated with sepsis.

19. The method of claim 1, wherein the method further comprises:

identifying information about the illness based on the analysis.

20. The method of claim 18 wherein the illness is sepsis, and the information includes identifying the risk that the sepsis is that of a gram-positive organism or gram-negative organism.

21. The method of claim 1, wherein method comprises:

choosing from a plurality of templates that corresponds to the pattern of the time-series data.

22. The method of claim 21, wherein the method further includes:

applying each of the plurality of templates individually and,

applying each of the plurality of templates for every possible combination of the plurality of templates that fit to the time-series data; and

choosing the individual template, or the combination of templates, that maximizes a predetermined variable.

23. A method comprising:

accepting time series data, said data of a cardiac signal including a heart beat or an RR interval;

analyzing the time series data using a pattern matching algorithm to identify pathological fluctuations in the signal; and

identifying a risk of clinical condition based on the analysis.

24. A system for identifying or monitoring a risk of a clinical condition, the system comprising at least one computer, a processor, at least one storage device, and at least one computer readable medium storing thereon a program, wherein the system comprises:

an input for accepting time series data, said data of a cardiac signal including a heart beat or an RR interval;

and the program is configured to, when executed by the processor, cause the system, to at least;

accept time series data of a cardiac signal;

obtain a pattern for the time series data;

choose a template function that corresponds to the pattern of the time-series data;

perform an analysis of the time series data to match sequences in the time series data to the template function, wherein one or more of the sequences comprises fluctuations;

calculate one or more characteristics of the fluctuations based on the analysis; and

identify a risk of the clinical condition associated with one of more of the characteristics.

25. The system of claim 24, wherein the fluctuations include accelerations or decelerations.

26. The system of claim 24, wherein the template function is a wavelet function.

27. The system of claim 26, wherein the wavelet function is selected from a group consisting of exponential, Gaussian, Lorentzian, and χ

(n),

wherein χ

(n) is

28. The system of claim 27, wherein the template function is χ

(n).

29. The system of claim 24, wherein the analysis comprises a sliding-window analysis, the analysis including sweeping the template function through the time series data in sequence increments and calculating a correlation coefficient between the template function and any sequence increment, wherein the sequence increment may comprise one or more fluctuations.

30. The system of claim 29, wherein different analytic forms of the template function are swept through the time series data.

31. The system of claim 30, wherein the different analytic forms of the template function that are swept through the time series data vary in amplitude, duration, or a combination thereof.

32. The system of claim 31, wherein the different analytic forms of the template function that is swept through the time series data increase in duration from about 8 to about 100 beats.

33. The system of claim 30, wherein the system is configured to at least:

determine the different analytic forms of the template function that highly correlate with the sequence increments.

34. The system of claim 33, wherein the correlation between the different analytic forms of the template function and the sequence increments is at a minimum threshold.

35. The system of claim 33, wherein the computer program is configured to cause the system to at least:

remove overlapping fluctuations.

36. The system of claim 35, wherein removal of overlapping fluctuations comprises identifying and keeping the overlapping fluctuation that has the highest correlation coefficient.

37. The system of claim 24, wherein the system is configured to at least:

remove the fluctuations from the cardiac signal and determining baseline signal variability.

38. The system of claim 37, wherein baseline signal variability is measured through sample asymmetry, smoothness, residual variability, or a combination thereof.

39. The system of claim 24, wherein the characteristics of the fluctuations are selected from the group consisting of number of fluctuations;

amplitude of the fluctuations;

widths of the fluctuations;

R² between the template function and fluctuation; and

baseline signal variability.

40. The system of claim 24, wherein the clinical condition is an illness.

41. The system of claim 40, wherein the illness is sepsis or conditions associated with sepsis.

42. The system of claim 24, wherein the system is configured to at least:

identify information about the illness based on the analysis.

43. The system of claim 42 wherein the illness is sepsis, and the information includes identifying the risk that the sepsis is that of a gram-positive organism or gram-negative organism.

44. The system of claim 24, wherein the system is configured to cause the system to identify a risk of the illness occurring in the subject.

45. The system of claim 24, wherein the system is configured to cause the system to:

choose from a plurality of templates that corresponds to the pattern of the time-series data.

46. The system of claim 45, wherein the system is configured to cause the system to:

apply each of the plurality of templates individually and,

apply each of the plurality of templates for every possible combination of the plurality of templates that fit to the time-series data; and

choosing the individual template, or the combination of templates, that maximizes a predetermined variable.

47. The system of claim 24, wherein the system is operatively connected to a cardiac signal monitoring system.

48. A system for monitoring a risk of a clinical condition, the system comprising at least one computer, a processor, at least one storage device, at least one computer readable medium storing thereon a program, wherein the system comprises:

an input for accepting time series data, said data of a cardiac signal including a heart beat or an RR interval; and

the program is configured to, when executed by the processor, cause the system to at least;

analyze the time series data using a pattern matching algorithm to identify pathological fluctuations in the signal; and

identify a risk of clinical condition based on the analysis.

49. A computer program product comprising a computer usable medium, comprising at least one computer, a processor, at least one storage device, and the computer readable medium storing thereon the program, wherein the program is configured to, when executed by the processor, cause a system to at least:

analyze time series data of a cardiac signal including a heart beat or an RR interval using a pattern matching algorithm to identify pathological fluctuations in the signal; and

identify a risk of clinical condition based on the analysis.