Apparatus for detecting sleep apnea using electrocardiogram signals

• US 7,025,729 B2
• Filed: 09/14/2001
• Issued: 04/11/2006
• Est. Priority Date: 09/14/2001
• Status: Expired due to Term

• Current Assignee: ResMed Sensor Technologies Limited (ResMed Incorporated)
• Original Assignee: BiancaMed Limited (ResMed Incorporated)
• Inventors: Heneghan, Conor, Sheridan, Elaine, de Chazal, Philip
• Primary Examiner(s): EVANISKO, GEORGE ROBERT
• Application Number:
  US09/952,688
• Publication Number:
  US 20030055348A1
• Time in Patent Office: 1,670 Days
• Field of Search: 600509-519
• US Class Current: 600/508
• CPC Class Codes: A61B 5/335 using integrated circuit memory devices
• CPC Class Codes: A61B 5/352 Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval
• CPC Class Codes: A61B 5/4818 Sleep apnoea
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/7207 of noise induced by motion artifacts
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

There is provided a method of determining a diagnostic measure of sleep apnea including the following steps: acquiring an electrocardiogram signal, calculating a set of RR intervals and electrocardiogram-derived respiratory signal from said electrocardiogram, and hence calculating a set of spectral and time-domain measurements over time periods including power spectral density, mean, and standard deviation. These measurements are processed by a classifier model which has been trained on a pre-existing data base of electrocardiogram signals to provide a probability of a specific time period containing apneic episodes or otherwise. These probabilities can be combined to form an overall diagnostic measure. The system also provides a system and apparatus for providing a diagnostic measure of sleep apnea.

Claims

1. An apparatus useful in diagnosing sleep apnea in a human patient, the apparatus comprising:

means for measuring a cardiac electrical potential of the patient and for generating an electrocardiogram signal over a time period;

means for analyzing the electrocardiogram signal and for deriving a respiratory signal therefrom;

means for partitioning the electrocardiogram signal into a set of time periods shorter than the time period of the electrocardiogram signal;

means for classifying each time period in the set of time periods as either apneic or normal; and

means for combining classification results from a plurality of the time periods and for providing a diagnostic measure of sleep apnea for the human patient based on the combined classification results.

2. The apparatus of claim 1, wherein the means for classifying comprises calculating, for each time period in the set of time periods, an associated probability that said each time period is apneic.

3. The apparatus of claim 1, wherein the means for analyzing the electrocardiogram signal comprises means for determining a sequence of flit intervals.

4. The apparatus of claim 3, further comprising means for correcting physiologically unreasonable RR intervals contained in the sequence of RR intervals.

5. The apparatus of claim 1, wherein the means for analyzing the electrocardiogram signal derives the respiratory signal by calculating an area enclosed by the electrocardiogram signal in a region around an R waveform peak.

6. The apparatus of claim 1, further comprising signal processing means for removing interference from the electrocardiogram signal and for producing a processed electrocardiogram signal which is provided to the means for analyzing the electrocardiogram signal.

7. The apparatus of claim 6, further comprising means for recording the processed electrocardiogram signal.

8. The apparatus of claim 1, wherein the means for combining classification results provides a diagnostic measure of sleep apnea for the human patient as at least one of an Apnea-Hyponea Index and a Respiratory Disturbance Index.

9. The apparatus of claim 1, wherein the means for analyzing the electrocardiogram signal comprises a computer software program executed by a processor.

10. The apparatus of claim 1, wherein the means for combining classification results and for providing a diagnostic measure of sleep apnea for the human patient comprises a display interface which provides a visual indication of a diagnosis.

11. The apparatus of claim 10, wherein the visual indication of the diagnosis comprises a color-coded indication.

12. The apparatus of claim 1, further comprising means for transmitting the electrocardiogram signal to a physical location remote from a location of the patient,

wherein the means for analyzing the electrocardiogram signal and for deriving a respiratory signal comprise processing means located at the remote physical location.

13. The apparatus of claim 1, wherein only a single channel of a cardiac electrical potential of the patient is analyzed by the means for analyzing the electrocardiogram signal.

14. The apparatus of claim 1, further comprising means for determining the corrected set of RR intervals for each time period in the set of time periods,

wherein the means for classifying each time period in the set of time periods uses a corrected set of RR intervals and the derived respiratory signal, and

wherein the classification results from a plurality of the time periods are combined to provide a diagnosis of apnea for the patient.

15. The apparatus of claim 1, wherein the means for partitioning and the means for analyzing the electrocardiogram signal create a measurement vector for each time period of the set of time periods,

wherein the measurement vector comprises spectral information from each of the derived respiratory signal and an RR sequence.

16. The apparatus of claim 1, wherein the means for classifying comprises using a classifier model which has been trained on a pro-existing database of signals.

17. The apparatus of claim 16, wherein the classifier model is a linear discriminant classifier.

18. The apparatus of claim 16, wherein the classifier model is a quadratic discriminant classifier.

19. The apparatus of claim 1, wherein the means for analyzing the electrocardiogram signal uses statistical parameters selected from the group consisting of:

a power spectral density of the electrocardiogram-derived respiratory signal;

a mean and a standard deviation of the electrocardiogram-derived respiratory signal;

a power spectral density of plural RR intervals;

a mean and standard deviation of plural RR intervals;

serial correlation coefficients of plural RR intervals;

an Allan factor A(T) evaluated at a time scale T of 5, 10, 15, 20 and 25 seconds, wherein A(T)=E{[Ni+1(T)−

Ni(T)]2}/2E{Ni+1(T)}, and where Ni(T) is a number of R wave peaks occurring in a window of length T stretching from iT to (i+1)T, and E is an expectation operator;

a NN50 measure (variant 1), defined as a number of pairs of adjacent RR intervals where a first interval exceeds a second interval by more than 50 ms;

a NN50 measure (variant 2), defined as a number of pairs of adjacent RR intervals where a second interval exceeds a first interval by more than 50 ms;

two pNN50 measures, defined as each NN50 measure divided by a total number of RR intervals;

SDSD measures, defined as a standard deviation of differences between adjacent RR intervals,

an RMSSD measure defined as a square root of a mean of a sum of squares of differences between adjacent RR intervals;

a mean or a standard deviation of an RR signal over an entire recorded electrocardiogram; and

a mean or a standard deviation of the electrocardiogram-derived respiratory signal over an entire recorded electrocardiogram.

20. The apparatus of claim 1, further comprising means for determining a quality of the generated electrocardiogram signal and for indicating whether the electrocardiogram signal is unsuitable for further processing.

21. The apparatus of claim 1, further comprising means for calculating a power spectral density of the electrocardiogram-derived respiratory signal using an averaged periodogram technique.

22. The apparatus of claim 1, wherein the means for analyzing the electrocardiogram signal comprises analyzing the derived respiratory signal.

23. The apparatus of claim 1, wherein the means for analyzing the electrocardiogram signal comprises means for determining a sequence of cardiac interbeat intervals.

24. The apparatus of claim 1, wherein the means for classifying each time period in the set of time periods as either apneic or normal comprises using the derived respiratory signal and the electrocardiogram signal.

25. An apparatus useful in diagnosing sleep apnea in a human patient, the apparatus comprising:

means for measuring a cardiac electrical potential of the patient and for generating an electrocardiogram signal over a time period;

means for analyzing the electrocardiogram signal comprising both time domain and frequency domain processing;

means for partitioning the electrocardiogram signal into a set of time periods shorter than the time period of the electrocardiogram signal;

means for classifying each time period in the set of time periods as either apneic or normal; and

means for combining classification results from a plurality of the time periods and for providing a diagnostic measure of sleep apnea for the human patient based on the combined classification results.

26. The apparatus of claim 25, wherein the means for classifying each time period in the set of rime periods as either apneic or normal comprises deriving and using a respiratory signal derived from the electrocardiogram signal.

27. An apparatus useful in diagnosing sleep apnea in a human patient, the apparatus comprising:

means for measuring a cardiac electrical potential of the patient and for generating an electrocardiogram signal over a time period;

means for analyzing the electrocardiogram signal and for deriving a respiratory signal therefrom;

means for partitioning the electrocardiogram signal into a set of time periods shorter than the time period of the electrocardiogram signal;

means for classifying each time period in the set of time periods as either apneic or normal; and

means for combining classification results from a plurality of the time periods and for providing a diagnostic measure of sleep apnea for the human patient based on the combined classification results,

wherein the means for analyzing the electrocardiogram signal comprises means for calculating a power spectral density of the electrocardiogram-derived respiratory signal.

28. The apparatus of claim 27, wherein the means for classifying each time period in the set of time periods as either apneic or normal comprises using the derived respiratory signal and the electrocardiogram signal.

29. An apparatus useful in diagnosing sleep apnea in a human patient, the apparatus comprising:

means for measuring a cardiac electrical potential of the patient and for generating an electrocardiogram signal over a time period;

means for analyzing the electrocardiogram signal and for deriving a respiratory signal therefrom;

means for partitioning the electrocardiogram signal into a set of time periods shorter than the time period of the electrocardiogram signal;

means for classifying each time period in the set of time periods as either apneic or normal; and

means for combining classification results from a plurality of the time periods and for providing a diagnostic measure of sleep apnea for the human patient based on the combined classification results;

means for determining a corrected set of RR intervals for each time period in the set of time periods,

wherein the means for classifying each time period in the set of time periods uses the corrected set of RR intervals and the derived respiratory signal,

wherein a diagnostic measure of sleep apnea for the patient is provided using a combined classification result from a plurality of the time periods.

30. The apparatus of claim 29, wherein the means for classifying each time period in the set of time periods as either apneic or normal comprises using the derived respiratory signal and the electrocardiogram signal.

31. An apparatus useful in diagnosing sleep apnea in a human patient, the apparatus comprising:

means for measuring a cardiac electrical potential of the patient and for generating an electrocardiogram signal over a time period;

means for analyzing the electrocardiogram signal and for partitioning the time period into plural time intervals each shorter than the time period of the electrocardiogram signal;

means for determining a power spectral density during each of the plural time intervals;

means for classifying each of the plural time intervals as either apneic or normal; and

means for combining classification results from each of the plural time intervals and for providing a diagnostic measure of sleep apnea for the human patient based on the combined classification results,

wherein the means for classifying each of the plural time intervals at least uses a measurement of a power spectral density of an associated one of the plural RR intervals.

32. The apparatus of claim 31, wherein the means for classifying each of the plural time intervals as either apneic or normal comprises deriving and using a respiratory signal derived from the electrocardiogram signal.

33. A method for diagnosing sleep apnea in a human patient, the comprising:

measuring a cardiac electrical potential of the patient and for generating an electrocardiogram signal over a time period;

analyzing the electrocardiogram signal and for deriving a respiratory signal therefrom;

partitioning the electrocardiogram signal into a set of time periods shorter than the time period of the electrocardiogram signal;

classifying each time period in the set of time periods as either apneic or normal; and

combining classification results from a plurality of the time periods and for providing a diagnostic measure of sleep apnea for the human patient based on the combined classification results.

34. The method of claim 33, wherein said classifying comprises calculating, for each time period in the set of time periods, an associated probability that said each time period is apneic.

35. The method of claim 33, wherein said analyzing the electrocardiogram signal comprises determining a sequence of RR intervals.

36. The method of claim 35, further comprising correcting physiologically unreasonable RR intervals contained in the sequence of RR intervals.

37. The method of claim 33, wherein said analyzing the electrocardiogram signal comprises deriving the respiratory signal by calculating an area enclosed by the electrocardiogram signal in a region around an R waveform peak.

38. The method of claim 33, farther comprising removing interference from the electrocardiogram signal and producing a processed electrocardiogram signal which is used to analyze the electrocardiogram signal.

39. The method of claim 38, further comprising recording the processed electrocardiogram signal.

40. The method of claim 33, wherein said combining classification results provides a diagnostic measure of sleep apnea for the human patient as at least one of an Apnea-Hyponea Index and a Respiratory Disturbance Index.

41. The method of claim 33, wherein said analyzing the electrocardiogram signal comprises using a computer software program executed by a processor.

42. The method of claim 33, wherein said combining classification results and providing a diagnostic measure of sleep apnea for the human patient comprises providing a display interface which provides a visual indication of a diagnosis.

43. The method of claim 42, wherein the visual indication of the diagnosis comprises a color-coded indication.

44. The method of claim 33, further comprising transmitting the electrocardiogram signal to a physical location remote from a location of the patient,

wherein said analyzing the electrocardiogram signal and deriving a respiratory signal comprise providing a processor located at the remote physical location.

45. The method of claim 33, wherein said analyzing the electrocardiogram signal analyzes only a single channel of a cardiac electrical potential of the patient.

46. The method of claim 33, further comprising determining a corrected set of RR intervals for each time period in the set of time periods,

wherein said classifying each time period in the set of time periods uses a corrected set of RR intervals and the derived respiratory signal, and

wherein the classification results from a plurality of the time periods are combined to provide a diagnosis of apnea for the patient.

47. The method of claim 33, wherein said partitioning and said analyzing the electrocardiogram signal create a measurement vector for each time period of the set of time periods,

wherein the measurement vector comprises spectral information from each of the derived respiratory signal and an RR sequence.

48. The method of claim 33, wherein said classifying comprises using a classifier model which has been trained on a pre-existing database of signals.

49. The method of claim 48, wherein the classifier model is a linear discriminant classifier.

50. The method of claim 48, wherein the classifier model is a quadratic discriminant classifier.

51. The method of claim 33, wherein said analyzing the electrocardiogram signal comprises using statistical parameters selected from the group consisting of:

a power spectral density of the electrocardiogram-derived respiratory signal;

a mean and a standard deviation of the electrocardiogram-derived respiratory signal;

a power spectral density of plural RR intervals;

a mean mid standard deviation of plural RR intervals;

serial correlation coefficients of plural RR intervals;

an Allan factor A(T) evaluated at a time scale T of 5, 10, 15, 20 and 25 seconds, wherein A(T)=E{[Ni+1(T)−

Ni(T)]2}/2E{Ni+1 (T)}, and where Ni(T) is a number of R wave peaks occurring in a window of length T stretching from iT to (i+1)T, and E is an expectation operator;

a NN50 measure (variant 1), defined as a number of pairs of adjacent RR intervals where a first interval exceeds a second interval by more than 50 ms;

a NN50 measure (variant 2), defined as a number of pairs of adjacent RR intervals where a second interval exceeds a first interval by more than 50 ms;

two pNN50 measures, defined as each NN50 measure divided by a total number of RR intervals;

SDSD measures, defined as a standard deviation of differences between adjacent RR intervals,

an RMSSD measure defined as a square root of a mean of a sum of squares of differences between adjacent RR intervals;

a mean or a standard deviation of an RR signal over an entire recorded electrocardiogram; and

a mean or a standard deviation of the electrocardiogram-derived respiratory signal over an entire recorded electrocardiogram.

52. The method of claim 33, further comprising determining a quality of the generated electrocardiogram signal and indicating whether the electrocardiogram signal is unsuitable for further processing.

53. The method of claim 33, further comprising calculating a power spectral density of the electrocardiogram-derived respiratory signal using an averaged periodogram technique.

54. The method of claim 33, wherein said analyzing the electrocardiogram signal comprises analyzing the derived respiratory signal.

55. The method of claim 33, wherein said analyzing the electrocardiogram signal comprises determining a sequence of cardiac interbeat intervals.

56. The method of claim 33, wherein said classifying each time period in the set of time periods as either apneic or normal comprises using the derived respiratory signal and the electrocardiogram signal.

57. A method useful in diagnosing sleep apnea in a human patient, the method comprising:

measuring a cardiac electrical potential of the patient and generating an electrocardiogram signal over a time period;

analyzing the electrocardiogram signal using both time domain and frequency domain processing;

partitioning the electrocardiogram signal into a set of time periods shorter than the time period of the electrocardiogram signal;

classifying each time period in the set of time periods as either apneic or normal; and

combining classification results from a plurality of the time periods and providing a diagnostic measure of sleep apnea for the human patient based on the combined classification results.

58. The method of claim 57, wherein said classifying each time period in the set of time periods as either apneic or normal comprises deriving and using a respiratory signal derived from the electrocardiogram signal.

59. A method useful in diagnosing sleep apnea in a human patient, the method comprising:

measuring a cardiac electrical potential of the patient and generating an electrocardiogram signal over a time period;

analyzing the electrocardiogram signal and deriving a respiratory signal therefrom;

partitioning the electrocardiogram signal into a set of time periods shorter than the time period of the electrocardiogram signal;

classifying each time period in the set of time periods as either apneic or normal; and

combining classification results from a plurality of the time periods and providing a diagnostic measure of sleep apnea for the human patient based on the combined classification results,

wherein said analyzing the electrocardiogram signal comprises calculating a power spectral density of the electrocardiogram-derived respiratory signal.

60. The method of claim 59, wherein said classifying each time period in the set of time periods as either apneic or normal comprises using the derived respiratory signal and the electrocardiogram signal.

61. A method useful in diagnosing sleep apnea in a human patient, the method comprising:

measuring a cardiac electrical potential of the patient and generating an electrocardiogram signal over a time period;

analyzing the electrocardiogram signal and deriving a respiratory signal therefrom;

partitioning the eelectrocardiogram signal into a set of time periods shorter than the time period of the electrocardiogram signal;

classifying each time period in the set of time periods as either apneic or normal; and

combining classification results from a plurality of the time periods and providing a diagnostic measure of sleep apnea for the human patient based on the combined classification results;

determining a corrected set of RR intervals for each time period in the set of time periods,

wherein said classifying each time period in the set of time periods uses the corrected set of RR intervals and the derived respiratory signal,

wherein a diagnostic measure of sleep apnea for the patient is provided using a combined classification result from a plurality of the time periods.

62. The method of claim 61, wherein said classifying each time period in the set of time periods as either apneic or normal comprises using the derived respiratory signal and the electrocardiogram signal.

63. A method useful in diagnosing sleep apnea in a human patient, the method comprising:

measuring a cardiac electrical potential of the patient and generating an electrocardiogram signal over a time period;

analyzing the electrocardiogram signal and partitioning the time period into plural rime intervals each shorter than the time period of the electrocardiogram signal;

determining a power spectral density during each of the plural time intervals;

classifying each of the plural time intervals as either apneic or normal; and

combining classification results from each of the plural time intervals and providing a diagnostic measure of sleep apnea for the human patient based on the combined classification results,

wherein said classifying each of the plural time intervals at least uses a measurement of a power spectral density of an associated one of the plural RR intervals.

64. The method of claim 63, wherein said classifying each of the plural time intervals as either apneic or normal comprises deriving and using a respiratory signal derived from the electrocardiogram signal.