Method for diagnosing heart disease, predicting sudden death, and analyzing treatment response using multifractial analysis

• US 20030163057A1
• Filed: 02/22/2002
• Published: 08/28/2003
• Est. Priority Date: 02/22/2002
• Status: Active Grant

• Current Assignee: Jacob Joseph, James T. Flick
• Original Assignee: Jacob Joseph, James T. Flick
• Inventors: Joseph, Jacob, Flick, James T.
• Granted Patent: US 6,993,377 B2
• Time in Patent Office: Days
• US Class Current: 600/509
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• 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: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 15/00 ICT specially adapted for medical reports, e.g. generation or transmission thereof
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders

Abstract

A method of analyzing electrocardiogram (EKG) data for use in the diagnosis of heart disease, prognosis of cardiac conditions, and the monitoring of heart disease therapies is disclosed. The method utilizes a wavelet-based multifractal analysis with one or more of (1) a discrete wavelet smoothing step to remove the effects of abnormal beats; (2) “Levy flight” analysis to detect the frequency of abnormal beats known to adversely affect the multifractal (MF) analysis; and (3) MF alpha analysis, a multifractal extension of monofractal short term (ST) alpha analysis. The invention further comprises an EKG test battery comprising Levy flight anomalous beat/beat cluster screening, followed by (ST) MF alpha analysis and MF Holder analysis (when validated by the Levy flight analysis). The wavelet smoothing step can also be used to classify human EKGs by observing the effect of sequential smoothing on the MF Holder coefficient. Alternative choices to the wavelet smoothing approach to removal of abnormal beat effects include probability distribution function analysis to determine the MF Holder coefficient directly, abnormal beat ridge skeleton removal to remove the offending beats based on a direct multifractal spectrum calculation, and the calculation of various types of entropy coefficients for the EKG time series.

Claims

1. A method of predicting a patient'"'"'s risk of sudden death from cardiac disease, comprising the steps of:

(a) collecting electrocardiogram data;

(b) smoothing the data by applying a wavelet transform filter to the data; and

(c) applying wavelet-based multifractal analysis to the data.

2. The method of claim 1, wherein the step of applying multifractal analysis to the data comprises the step of calculating a multifractal predictive coefficient, wherein the predictive coefficient has a numerical value that correlates to the patient'"'"'s risk of sudden death from cardiac disease.

3. The method of claim 2, wherein said step of calculating a multifractal coefficient comprises the step of calculating a multifractal Holder coefficient.

4. The method of claim 1, wherein said step of applying a wavelet transform filter comprises the step of applying a MODWT wavelet transform filter.

5. The method of claim 1, further comprising the step of performing a probability distribution on the data prior to the step of applying wavelet-based multifractal analysis.

6. The method of claim 1, further comprising the step of performing a cluster analysis to the data.

7. The method of claim 6, wherein said step of performing a cluster analysis comprises the step of performing a Levy flight cluster analysis.

8. The method of claim 1, further comprising the step of calculating a short-scale multifractal coefficient, wherein said short-scale multifractal coefficient characterizes the short-scale beat region of the data.

9. The method of claim 8, wherein said step of calculating a short-scale coefficient comprises the step of calculating a multifractal alpha coefficient.

10. The method of claim 9, wherein said multifractal alpha coefficient characterizes a scale of one of 1-4 to 1-16 beats.

11. The method of claim 1, further comprising the step of quantifying the multifractality of the data by generating a multifractal cascade coefficient.

12. The method of claim 11, wherein said step of quantifying the multifractality of the data includes the step of generating a quadratic coefficient, wherein the value of said quadratic coefficient is indicative of the degree of multifractality of the data.

13. The method of claim 12, wherein said quadratic coefficient is indicative of multifractal data when the quadratic coefficient has a value greater than a threshold value.

14. A method of determining the severity of cardiac disease, comprising the steps of:

(a) collecting electrocardiogram data;

(b) performing a probability distribution on the data; and

(c) applying wavelet-based multifractal analysis to the data.

15. The method of claim 14, wherein the step of applying multifractal analysis to the data comprises the step of calculating a multifractal predictive coefficient, wherein the predictive coefficient has a numerical value that correlates to the patient'"'"'s risk of sudden death from cardiac disease.

16. The method of claim 15, wherein said step of calculating a multifractal coefficient comprises the step of calculating a multifractal Holder coefficient.

17. The method of claim 14, wherein said step of applying a wavelet transform filter comprises the step of applying a MODWT wavelet transform filter.

18. The method of claim 14, further comprising the step of applying a wavelet transform filter to the data prior to the step of applying wavelet-based multifractal analysis.

19. The method of claim 14, further comprising the step of performing a cluster analysis to the data.

20. The method of claim 19, wherein said step of performing a cluster analysis comprises the step of performing a Levy flight cluster analysis.

21. The method of claim 14, further comprising the step of calculating a short-scale multifractal coefficient, wherein said short-scale multifractal coefficient characterizes the short-scale beat region of the data.

22. The method of claim 21, wherein said step of calculating a short-scale coefficient comprises the step of calculating a multifractal alpha coefficient.

23. The method of claim 22, wherein said multifractal alpha coefficient characterizes a scale of one of 1-8 to 1-16 beats.

24. The method of claim 14, further comprising the step of quantifying the multifractality of the data by generating a multifractal cascade coefficient.

25. The method of claim 24, wherein said step of quantifying the multifractality of the data includes the step of generating a quadratic coefficient, wherein the value of said quadratic coefficient is indicative of the degree of multifractality of the data.

26. The method of claim 25, wherein said quadratic coefficient is indicative of multifractal data when the quadratic coefficient has a value greater than a specified threshold.

27. A method of diagnosing cardiac disease in a patient, comprising the steps of:

(a) collecting electrocardiogram data;

(b) smoothing the data by applying a wavelet transform filter to the data; and

(c) applying wavelet-based multifractal analysis to the data.

28. The method of claim 27, wherein the step of applying multifractal analysis to the data comprises the step of calculating a multifractal predictive coefficient, wherein the predictive coefficient has a numerical value that correlates to the presence of cardiac disease in the patient.

29. The method of claim 28, wherein said step of calculating a multifractal coefficient comprises the step of calculating a multifractal Holder coefficient.

30. The method of claim 27, wherein said step of applying a wavelet transform filter comprises the step of applying a MODWT wavelet transform filter.

31. The method of claim 27, further comprising the step of performing a probability distribution on the data prior to the step of applying wavelet-based multifractal analysis.

32. The method of claim 27, further comprising the step of performing a cluster analysis to the data.

33. The method of claim 32, wherein said step of performing a cluster analysis comprises the step of performing a Levy flight cluster analysis.

34. The method of claim 27, further comprising the step of calculating a short-scale multifractal coefficient, wherein said short-scale multifractal coefficient characterizes the short-scale beat region of the data.

35. The method of claim 34, wherein said step of calculating a short-scale coefficient comprises the step of calculating a multifractal alpha coefficient.

36. The method of claim 35, wherein said multifractal alpha coefficient characterizes a scale range of 1-4 to 1-16 beats.

37. The method of claim 27, further comprising the step of quantifying the multifractality of the data by generating a multifractal cascade coefficient.

38. The method of claim 37, wherein said step of quantifying the multifractality of the data includes the step of generating a quadratic coefficient, wherein the value of said quadratic coefficient is indicative of the degree of multifractality of the data.

39. The method of claim 38, wherein said quadratic coefficient is indicative of multifractal data when the quadratic coefficient has a value greater than a threshold value.

40. A method of diagnosing cardiac disease in a patient, comprising the steps of:

(a) collecting electrocardiogram data;

(b) performing a probability distribution on the data; and

(c) applying wavelet-based multifractal analysis to the data.

41. The method of claim 40, wherein the step of applying multifractal analysis to the data comprises the step of calculating a multifractal predictive coefficient, wherein the predictive coefficient has a numerical value that correlates to the presence of cardiac disease in the patient.

42. The method of claim 41, wherein said step of calculating a multifractal coefficient comprises the step of calculating a multifractal Holder coefficient.

43. The method of claim 40, wherein said step of applying a wavelet transform filter comprises the step of applying a MODWT wavelet transform filter.

44. The method of claim 40, further comprising the step of applying a wavelet transform filter to the data prior to the step of applying wavelet-based multifractal analysis.

45. The method of claim 40, further comprising the step of performing a cluster analysis to the data.

46. The method of claim 45, wherein said step of performing a cluster analysis comprises the step of performing a Levy flight cluster analysis.

47. The method of claim 40, further comprising the step of calculating a short-scale multifractal coefficient, wherein said short-scale multifractal coefficient characterizes the short-scale beat region of the data.

48. The method of claim 47, wherein said step of calculating a short-scale coefficient comprises the step of calculating a multifractal alpha coefficient.

49. The method of claim 48, wherein said multifractal alpha coefficient characterizes a scale of one of 1-8 and 1-16 beats.

50. The method of claim 40, further comprising the step of quantifying the multifractality of the data by generating a multifractal cascade coefficient.

51. The method of claim 50, wherein said step of quantifying the multifractality of the data includes the step of generating a quadratic coefficient, wherein the value of said quadratic coefficient is indicative of the degree of multifractality of the data.

52. The method of claim 51, wherein said quadratic coefficient is indicative of multifractal data when the quadratic coefficient has a value of about 11 or greater.

53. A method of evaluating a treatment regimen for heart disease, comprising the steps of:

(a) collecting electrocardiogram data;

(b) smoothing the data by applying a wavelet transform filter to the data; and

(c) applying wavelet-based multifractal analysis to the data.

54. The method of claim 53, wherein the step of applying multifractal analysis to the data comprises the step of calculating a multifractal predictive coefficient, wherein the predictive coefficient has a numerical value that correlates to the effectiveness of the treatment regimen.

55. The method of claim 54, wherein said step of calculating a multifractal coefficient comprises the step of calculating a multifractal Holder coefficient.

56. The method of claim 53, wherein said step of applying a wavelet transform filter comprises the step of applying a MODWT wavelet transform filter.

57. The method of claim 53, further comprising the step of performing a probability distribution on the data prior to the step of applying wavelet-based multifractal analysis.

58. The method of claim 53, further comprising the step of performing a cluster analysis to the data.

59. The method of claim 58, wherein said step of performing a cluster analysis comprises the step of performing a Levy flight cluster analysis.

60. The method of claim 53, further comprising the step of calculating a short-scale multifractal coefficient, wherein said short-scale multifractal coefficient characterizes the short-scale beat region of the data.

61. The method of claim 60, wherein said step of calculating a short-scale coefficient comprises the step of calculating a multifractal alpha coefficient.

62. The method of claim 61, wherein said multifractal alpha coefficient characterizes a scale of one of 1-8 and 1-16 beats.

63. The method of claim 53, further comprising the step of quantifying the multifractality of the data by generating a multifractal cascade coefficient.

64. The method of claim 64, wherein said step of quantifying the multifractality of the data includes the step of generating a quadratic coefficient, wherein the value of said quadratic coefficient is indicative of the degree of multifractality of the data.

65. The method of claim 64, wherein said quadratic coefficient is indicative of multifractal data when the quadratic coefficient has a value of about 11 or greater.

66. A method of evaluating a treatment regimen for heart disease, comprising the steps of:

(a) collecting electrocardiogram data;

(b) performing a probability distribution on the data; and

(c) applying wavelet-based multifractal analysis to the data.

67. The method of claim 66, wherein the step of applying multifractal analysis to the data comprises the step of calculating a multifractal predictive coefficient, wherein the predictive coefficient has a numerical value that correlates to the effectiveness of the treatment regimen.

68. The method of claim 67, wherein said step of calculating a multifractal coefficient comprises the step of calculating a multifractal Holder coefficient.

69. The method of claim 66, wherein said step of applying a wavelet transform filter comprises the step of applying a MODWT wavelet transform filter.

70. The method of claim 66, further comprising the step of applying a wavelet transform filter to the data prior to the step of applying wavelet-based multifractal analysis.

71. The method of claim 66, further comprising the step of performing a cluster analysis to the data.

72. The method of claim 71, wherein said step of performing a cluster analysis comprises the step of performing a Levy flight cluster analysis.

73. The method of claim 66, further comprising the step of calculating a short-scale multifractal coefficient, wherein said short-scale multifractal coefficient characterizes the short-scale beat region of the data.

74. The method of claim 73, wherein said step of calculating a short-scale coefficient comprises the step of calculating a multifractal alpha coefficient.

75. The method of claim 74, wherein said multifractal alpha coefficient characterizes a scale of one of 1-8 and 1-16 beats.

76. The method of claim 66, further comprising the step of quantifying the multifractality of the data by generating a multifractal cascade coefficient.

77. The method of claim 76, wherein said step of quantifying the multifractality of the data includes the step of generating a quadratic coefficient, wherein the value of said quadratic coefficient is indicative of the degree of multifractality of the data.

78. The method of claim 77, wherein said quadratic coefficient is indicative of multifractal data when the quadratic coefficient has a value greater than a threshold value.

79. A method for determining one of whether a patient has heart disease, a patient'"'"'s risk of sudden death from a cardiac event, and the efficacy of a cardiac disease therapy, comprising the steps of:

(a) collecting EKG time series data for the patient;

(b) perform a cluster analysis on the data; and

(c) performing a multifractal alpha analysis on the data.

80. The method of claim 79, wherein said step of performing a cluster analysis comprises the step of performing a Levy flight analysis on the data.

81. The method of claim 80, wherein said step of performing a Levy flight analysis further comprises the step of determining a beat cluster index for the data.

82. The method of claim 79, wherein said step of performing a multifractal alpha analysis comprises the step of calculating an multifractal alpha coefficient for the data.

83. The method of claim 82, wherein said step of performing a multifractal alpha analysis further comprises the steps of calculating a critical multifractal alpha value and comparing the multifractal alpha coefficient to the critical multifractal alpha value.

84. The method of claim 79, further comprising the step of performing smoothing on the data.

85. The method of claim 84, wherein said smoothing step comprises the performance of wavelet smoothing on the data.

86. The method of claim 85, wherein said wavelet smoothing step comprises wavelet smoothing performed within a scaling range of 1-16 beats.

87. The method of claim 85, further comprising the step of calculating a multifractal cascade coefficient for the data.

88. The method of claim 87, further comprising the step of calculating a critical multifractal cascade value and comparing the multifractal cascade coefficient to the critical multifractal cascade value.

89. The method of claim 85, further comprising the step of performing multifractal Holder sequential smoothing on the data to generate a series of multifractal holder coefficients.

90. The method of claim 89, further comprising the step of determining if the multifractal holder coefficients increase over the course of the sequential smoothing step.

91. The method of claim 90, further comprising the step of calculating a critical multifractal Holder value and comparing the critical value to the multifractal holder coefficients.

92. The method of claim 84, wherein said smoothing step comprises the steps of generating a probability distribution function for the data wherein said probability distribution function has a head and a tail, and discarding at least some portion of the tail of the probability distribution function from the data.

93. The method of claim 84, wherein said smoothing step comprises the steps of generating wavelet ridges for the data and selectively removing at least some of the wavelet ridges that are related to intermittency in the data.

94. A method for determining one of whether a patient has heart disease, a patient'"'"'s risk of sudden death from a cardiac event, and the efficacy of a cardiac disease therapy, comprising the steps of:

(a) collecting EKG time series data for the patient;

(b) performing a direct multifractal spectrum calculation from the EKG time series data; and

(c) calculating one of a multifractal Holder analysis and a multifractal alpha analysis from the results of the step of performing a direct multifractal spectrum calculation.

96. The method of claim 94, wherein said step of calculating comprises the step of calculating a Tsallis entropic scaling coefficient according to the formula 1/Hmin−

1/Hmax.

95. A method for determining one of whether a patient has heart disease, a patient'"'"'s risk of sudden death from a cardiac event, and the efficacy of a cardiac disease therapy, comprising the steps of:

(a) collecting EKG time series data for the patient; and

(b) calculating one of Shannon entropy, Pesin metric and topological entropies, Kolmolgorov-Sinai entropy, approximate entropy, Lyapunov exponents, and a Tsallis entropic scaling coefficient from the EKG time series data.