System and method of monitoring physiological signals

• US 20070244402A1
• Filed: 02/20/2007
• Published: 10/18/2007
• Est. Priority Date: 02/17/2006
• Status: Abandoned Application

• Current Assignee: Data Sciences U K Limited (International Business Machines Corporation)
• Original Assignee: Data Sciences U K Limited (International Business Machines Corporation)
• Inventors: Brockway, Brian, Arnold, John, Ricci, Carlos, Mazar, Scott, Sherwood, Kathy, Kovtun, Vladimir, Jasperson, Keith
• Application Number:
  US11/677,015
• Publication Number:
  US 20070244402A1
• Time in Patent Office: Days
• US Class Current: 600/509
• CPC Class Codes: A61B 5/0006 ECG or EEG signals
• CPC Class Codes: A61B 5/361 Detecting fibrillation
• 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: A61N 1/3702 Physiological parameters A61N1/365 takes precedence; evoked response A61N1/371
• CPC Class Codes: G16H 40/63 for local operation
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment

Abstract

A physiological signal monitoring and analysis system including an implantable medical device and a signal processor. The implantable medical device is configured to monitor and record sample segments of at least one physiological signal of a patient at time separated recording intervals over a time period. The signal processor configured to measure values of at least one selected characteristic of the at least one physiological signal from the recorded sample segments, to determine trend information representing a trend in the at least one selected characteristic based on the measured values, and to assess a risk of a physiological event to the patient based on the trend information.

Claims

1. A physiological signal monitoring and analysis system comprising:

an implantable medical device configured to monitor and record sample segments of at least one physiological signal of a patient at time separated recording intervals over a time period; and

a signal processor configured to measure values of at least one selected characteristic of the at least one physiological signal from the recorded sample segments, to determine trend information representing a trend in the at least one selected characteristic based on the measured values, and to assess a risk of a physiological event to the patient based on the trend information.

2. The system of claim 1, wherein the signal processor is configured to determine the trend information by performing a modeling operation based on the measured values.

3. The system of claim 2, wherein the modeling operation is selected from the group consisting of curve fitting techniques, auto-regressive moving average (ARMA) techniques, auto-regressive integrated moving average (ARIMA) techniques, Kalman filtering techniques, and statistical techniques including mean, median, root-mean-square, trimmed mean, and moving-window statistics.

4. The system of claim 1, wherein the trend information comprises the measured values.

5. The system of claim 1, wherein the signal processor is configured to assess the risk by applying neural network techniques to the measured values.

6. The system of claim 1, wherein the signal processor is configured to assess the risk by assigning the patient to one of a plurality of risk stratification groups based on the trend information.

7. The system of claim 1, wherein the signal processor is configured to assess the risk by determining a risk value based on the trend information and to assign the patient to one of a plurality of risk stratification groups based on a comparison of the risk value to a risk decision value.

8. The system of claim 7, where the risk decision value comprises a risk decision value range.

9. The system of claim 7, wherein the signal processor is configured to determine the risk value by performing a mapping operation to map the trend information into a risk index.

10. The system of claim 9, wherein the risk index comprises a multivariate risk index.

11. The system of claim 10, wherein the mapping operation is selected from the group consisting of neural networking (NN) techniques, support vector machine (SVM) techniques, self-organizing mapping (SOM) techniques, generative topographic mapping (GTM) techniques, and fuzzy-neuro techniques.

12. The system of claim 1, wherein the signal processor is configured to assess the risk based on a deviation from a determined trend.

13. The system of claim 1, wherein the time separated recording intervals are selectable.

14. The system of claim 1, wherein the implantable medical device is configured to monitor a heart rate of the patient and to record a sample of the at least one physiological signal when the heart rate is within a predetermined heart rate range.

15. The system of claim 1, wherein the signal processor is configured to form a time-series of data values based on the measured values and to fit a curve to the time-series of data values, wherein the curve is representative of the trend in the at least one selected characteristic.

16. The system of claim 15, where the data values of the time-series comprise the measured values of the at least one selected characteristic.

17. The system of claim 15, wherein each of the data values of the time-series comprises a composite value of a corresponding selected plurality of the measured values of the at least one selected characteristic.

18. The system of claim 15, wherein the signal process is configured to extrapolate from the curve to estimate a future value of the time-series of values.

19. The system of claim 1, wherein the at least one selected characteristic comprises a plurality of characteristics, and wherein the signal processor is configured to measure values of each characteristic of the plurality of characteristics, to determine trend information for each characteristic of the plurality based on the corresponding measured values, and to assess the risk of the physiological event to the patient based on trend information of up to all characteristics of the plurality of characteristics.

20. The system of claim 19, wherein the signal processor is configured to employ multivariate analysis techniques to assess the risk.

21. The system of claim 1, wherein the at least one physiological signal comprises an electrocardiogram signal and the physiological event comprises cardiac arrhythmia.

22. The system of claim 1, wherein the at least one physiological signal comprises an electrocardiogram signal and the at least one selected characteristic is selected from the group consisting of:

T-wave amplitude alternans, T-wave area alternans, T-wave duration alternans, QT interval alternans, ST interval alternans, ST segment duration alternans, ST segment elevation alternans, RR interval alternans, R-wave amplitude alternans, R-wave area alternans, R-wave duration alternans, heart rate turbulence, QRS complex duration alternans, QRS complex area alternans, and QRS complex amplitude alternans.

23. A method of monitoring and analyzing at least one physiological signal of a patient, the method comprising:

recording sample segments of at least one physiological signal of a patient at time separated recording intervals over a time period using a medical device implanted within the patient;

measuring values of at least one selected characteristic of the at least one physiological signal from the recorded sample segments;

determining trend information representing a trend in the at least one selected characteristic based on the measured values; and

assessing a risk to the patient of a physiological event based on the trend information.

24. The method of claim 23, wherein determining the trend information includes performing a modeling operation based on the measured values.

25. The method of claim 24, wherein the modeling operation is selected from the group consisting of curve fitting techniques, auto-regressive moving average (ARMA) techniques, auto-regressive integrated moving average (ARIMA) techniques, and Kalman filtering techniques.

26. The method of claim 24, wherein the modeling operation is selected from a group of statistical techniques including mean, median, root-mean-square, trimmed mean, and moving-window statistics.

27. The method of claim 23, wherein the trend information comprises the measured values.

28. The method of claim 23, wherein assessing the risk includes applying neural network techniques to the measured values.

29. The method of claim 23, wherein assessing the risk includes assigning the patient to one of a plurality of risk stratification groups based on the trend information.

30. The method of claim 23, wherein assessing the risk includes:

determining a risk value based on the trend information;

comparing the risk value to a risk decision value; and

assigning the patient to one of a plurality of risk stratification groups based on the comparison.

31. The method of claim 30, wherein the risk decision value comprises a risk decision value range.

32. The method of claim30, wherein determining the risk value includes performing a mapping operation to map the trend information to a risk index.

33. The method of claim 32, wherein the mapping operation is selected from the group consisting of neural networking (NN) techniques, support vector machine (SVM) techniques, self-organizing mapping (SOM) techniques, generative topographic mapping (GTM) techniques, and fuzzy-neuro techniques.

34. The method of claim 32, wherein the risk index comprises a multivariate risk index.

35. The method of claim 23, wherein assessing the risk is based on a deviation from a determined trend.

36. The method of claim 23, wherein the time separated recording intervals are selectable.

37. The method of claim 23, wherein recording sample segments includes:

monitoring the patient'"'"'s heart rate; and

recording a sample segment of the at least one physiological signal when the heart rate is within a predetermined heart rate range.

38. The method of claim 23, wherein determining the trend includes:

forming a time-series of data values based on the measured values; and

fitting a curve to the time-series, wherein the fitted curve is representative of the trend of the at least one selected characteristic.

39. The method of claim 38, wherein the data values of the time-series comprise the measured values of the at least one selected characteristic.

40. The method of claim 38, wherein each of the data values of the time-series comprises a composite value of a corresponding selected plurality of the measured values of the at least one selected characteristic.

41. The method of claim 38, wherein determining the trend includes extrapolating from the fitted curve to estimate a future value of the time-series of values.

42. The method of claim 23, wherein the at least one selected characteristic comprises a plurality of characteristics, and wherein:

measuring values of the at least one selected characteristic includes measuring values of each of characteristics of the plurality of characteristics from the recorded samples;

determining the trend includes determining trend information representing each of the characteristics of the plurality of characteristics; and

assessing the risk includes correlating the trend information of up to all characteristics of the plurality of characteristics.

43. The method of claim 42, wherein correlating the trend information includes performing a multivariate analysis of the trend information of up to all characteristics of the plurality of characteristics.

44. The method of claim 23, wherein the at least one physiological signal comprises an electrocardiogram signal and the physiological event comprises a cardiac arrhythmia.

45. The method of claim 23, wherein the at least one selected characteristic comprises a T-wave alternans.

46. The method of claim 45, wherein the T-wave alternans is selected from the group consisting of:

T-wave amplitude alternans, T-wave area alternans, and T-wave duration alternans.

47. The method of claim 23, wherein the at least one selected characteristic comprises an electrocardiogram signal interval alternans.

48. The method of claim 47, where the electrocardiogram signal interval alternans is selected from the group consisting of:

QT interval alternans, ST interval alternans, RR interval alternans, heart rate turbulence, TT interval alternans, and PR interval alternans.

49. The method of claim 23, wherein the at least one selected characteristic comprises an R-wave alternans.

50. The method of claim 49, wherein the R-wave alternans is selected from the group consisting of:

R-wave amplitude alternans, R-wave area alternans, and R-wave duration alternans.

51. The method of claim 23, wherein the at least one selected characteristic comprises a QRS complex alternans.

52. The method of claim 51, wherein the QRS complex alternans is selected from the group consisting of:

QRS complex duration alternans and QRS complex area alternans.

53. The method of claim 23, wherein the at least one selected characteristic is selected from the group consisting of ST segment elevation.

54. The method of claim 23, wherein the recording intervals are time separated by 24-hours.

55. The method of claim 23, wherein the recording intervals are time separated by one week.

56. The method of claim 23, wherein the time period of one year.

57. A method of assessing the risk of a patient for sudden cardiac death, the method comprising:

recording sample segments of an electrocardiogram signal of the patient at time separated recording intervals over a time period using an implantable medical device implanted within the patient;

measuring values of at least one selected characteristic of the electrocardiogram signal from the recorded sample segments;

determining trend information representing a trend in the at least one selected characteristic based on the measured values; and

assessing a risk to the patient of suffering a cardiac arrhythmia leading to sudden cardiac death based on the trend information.

58. The method of claim 57, wherein determining the trend information includes performing a modeling operation based on the measured values.

59. The method of claim 58, wherein the modeling operation is selected from the group consisting of curve fitting techniques, auto-regressive moving average (ARMA) techniques, auto-regressive integrated moving average (ARIMA) techniques, and Kalman filtering techniques.

60. The method of claim 58, wherein the modeling operation is selected from a group of statistical techniques including mean, median, root-mean-square, trimmed mean, and moving-window statistics.

61. The method of claim 57, wherein the trend information comprises the measured values.

62. The method of claim 57, wherein assessing the risk includes applying neural networking techniques to the measured values.

63. The method of claim 57, wherein assessing the risk includes assigning the patient to one of a plurality of risk stratification groups based on the trend information.

64. The method of claim 57, wherein assessing the risk includes:

determining a risk value based on the trend information;

comparing the risk value to a risk decision value; and

assigning the patient to one of a plurality of sudden cardiac death risk stratification groups based on the comparison.

65. The method of claim 64, wherein the risk decision value comprises a risk decision value range.

66. The method of claim 64, wherein determining the risk value includes performing a mapping operation to map the trend information to a risk index comprising a plurality of risk values.

67. The method of claim 66, where the mapping operation is selected from the group consisting of neural networking (NN) techniques, support vector machine (SVM) techniques, self-organizing mapping (SOM) techniques, generative topographic mapping (GTM) techniques, and fuzzy-neuro techniques.

68. The method of claim 66, wherein the risk index comprises a multivariate risk index.

69. The method of claim 57, wherein assessing the risk is based on a deviation from a determined trend.

70. The method of claim 57, where the time separated recording intervals are selectable.

71. The method of claim 57, wherein recording the sample segments includes:

monitoring the heart rate of the patient; and

recording a sample segment of the electrocardiogram signal when the heart rate is within a predetermined heart rate range.

72. The method of claim 71, wherein the predetermined heart rate range is from 92 beats per minutes to 115 beats per minute inclusive.

73. The method of claim 57, wherein determining the trend information includes:

forming a time-series of data values based on the measured values; and

fitting a curve to the time-series, wherein the fitted curve is representative of the trend of the at least one selected characteristic.

74. The method of claim 73, wherein assessing the risk includes:

extrapolating from the fitted curve to determine a future data value of the time-series;

comparing the future data value to a risk decision value which is indicative of a risk level of suffering a cardiac arrhythmia.

75. The method of claim 73, wherein the data values of the time-series comprise the measured values of the at least one selected characteristic.

76. The method of claim 73, wherein each of the data values of the time-series comprise a composite value of a corresponding selected plurality of the measured values of the at least one selected characteristic.

77. The method of claim 57, wherein the at least one selected characteristic comprises a plurality of characteristics, and wherein:

measuring values of the at least one selected characteristic of the electrocardiogram signal includes measuring values of each of the characteristics of the plurality of characteristics from the recorded samples;

determining the trend includes determining trend information representing each of the characteristics of the plurality of characteristics; and

assessing the risk includes correlating the trend information of up to all characteristics of the plurality of characteristics.

78. The method of claim 77, wherein correlating the trend information includes performing a multivariate analysis of the trend information of the first and second characteristics.

79. The method of claim 57, wherein the at least one selected characteristic is selected from the group consisting of:

T-wave amplitude alternans, T-wave area alternans, T-wave duration alternans, QT interval alternans, ST interval alternans, ST segment duration alternans, ST segment elevation alternans, RR interval alternans, R-wave amplitude alternans, R-wave area alternans, R-wave duration alternans, heart rate turbulence, QRS complex duration alternans, QRS complex area alternans, and QRS complex amplitude alternans.