Method, system and apparatus for determining prognosis in atrial fibrillation

• US 6,064,906 A
• Filed: 03/11/1998
• Issued: 05/16/2000
• Est. Priority Date: 03/14/1997
• Status: Expired due to Fees

• Current Assignee: Emory University
• Original Assignee: Emory University
• Inventors: Bollmann, Andreas, Langberg, Jonathan J.
• Primary Examiner(s): Getzow, Scott M.
• Application Number:
  US09/038,615
• Time in Patent Office: 797 Days
• Field of Search: 600/515, 600/516, 600/517, 600/518
• US Class Current: 600/518
• CPC Class Codes: A61B 5/35 by template matching
• CPC Class Codes: A61B 5/361 Detecting fibrillation
• CPC Class Codes: A61B 5/7232 involving compression of the physiological signal, e.g. to extend the signal recording period
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• 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: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A system, apparatus and method for determining prognosis in atrial fibrillation non-invasively detects electrical activity of the heart with external electrodes, such as with ECG leads. The diagnostic system, apparatus and method provide a non-invasive technique to classify the severity of atrial fibrillation according to measured values of at least one predetermined parameter of the fibrillation as observed in the frequency domain. One predetermined parameter which is measured for the purpose of classification is peak frequency. Another predetermined parameter used for atrial fibrillation classification is a change in power along the range of frequencies. The system, apparatus and method use observed correlations between the measured parameter and success or failures with various treatment modalities in order to provide useful information for treatment of the atrial fibrillation, including the prediction of conversion with antiarrhthmic drug such as ibutilide, and the estimation of time to recurrence after cardioversion.

Claims

1. An atrial fibrillation classification apparatus, comprising:

a signal processor which is adapted to receive electrical signals indicative of a heart'"'"'s activity, to remove ventricular activity from the electrical signals, and to output a fibrillatory baseline signal;

a transformation circuit which is adapted to receive the fibrillatory baseline signal and to generate a set of frequency domain signals corresponding to the fibrillatory baseline signal; and

a detector which is adapted to detect a value of a predetermined parameter within the set of frequency domain signals;

wherein the signal processor is adapted to classify the atrial fibrillation based on the detected value of the predetermined parameter.

2. The apparatus as set forth in claim 1, further comprising an electrode which is adapted to detect the electrical signals indicative of the heart'"'"'s activity and to transmit the electrical signals to the signal processor.

3. The apparatus as set forth in claim 2, wherein the electrode is adapted to be placed on an external surface of a patient.

4. The apparatus as set forth in claim 2, wherein the electrode is adapted to be placed within a patient and to detect the fibrillatory baseline signal.

5. The apparatus as set forth in claim 1, further comprising an interface which is adapted to supply digital signals indicative of the heart'"'"'s activity to the signal processor.

6. The apparatus as set forth in claim 5, wherein the interface is adapted to receive the digital signals from a storage device.

7. The apparatus as set forth in claim 5, wherein the interface is adapted to receive the digital signals from a remote transmitter.

8. The apparatus as set forth in claim 1, further comprising an interface which is adapted to receive analog signals indicative of the heart'"'"'s electrical activity.

9. The apparatus as set forth in claim 8, wherein the interface is adapted to receive the analog signals from an ECG machine.

10. The apparatus as set forth in claim 1, wherein the signal processor is adapted to detect QRST complexes in the electrical signals, to generate one or more QRST templates, and to subtract QRST complexes using the one or more QRST templates.

11. The apparatus as set forth in claim 10, wherein each of the QRST complexes has a duration in time and the signal processor is adapted to generate the one or more templates by averaging the QRST complexes according to their duration in time.

12. The apparatus as set forth in claim 1, wherein the fibrillatory baseline signal comprises a digital signal and the signal processor is adapted to compress the fibrillatory baseline signal.

13. The apparatus as set forth in claim 12, wherein the signal processor is adapted to apply a Hamming window when compressing the fibrillatory baseline signal.

14. The apparatus as set forth in claim 1, wherein the transformation circuit is adapted to perform a Fourier transformation of the fibrillatory baseline signal to produce the set of frequency domain signals.

15. The apparatus as set forth in claim 1, further comprising an analog-to-digital converter which is adapted to convert the electrical signals into digital signals indicative of the heart'"'"'s activity.

16. The apparatus of claim 1, wherein

the detector further comprises a transition detector which is adapted to detect a change in power level within the set of frequency domain signals;

and

the signal processor is further adapted to classify the atrial fibrillation based upon the detected change in power level.

17. The apparatus as set forth in claim 16, wherein the signal processor is adapted to classify the atrial fibrillation by measuring a midpoint between mean values of high and low frequencies of the set of frequency domain signals.

18. The system as set forth in claim 17, wherein the signal processor is further adapted to measure a band of frequencies having an abrupt fall in power levels over a predetermined frequency range.

19. The system as set forth in claim 18, wherein the signal processor is further adapted to measure the abrupt fall in power levels over a frequency range of about 1 Hertz.

20. The system as et forth in claim 18, wherein the abrupt fall in power levels comprises a fall of over about one-half of an order of magnitude.

21. The apparatus as set forth in claim 17, wherein the signal processor is further adapted to indicate results after classifying the atrial fibrillation.

22. A method for classifying atrial fibrillation, comprising:

receiving electrical signals indicative of a heart'"'"'s activity;

processing the electrical signals to remove ventricular activity and to output a fibrillatory baseline signal;

transforming the fibrillatory baseline signal into a set of frequency domain signals;

detecting a predetermined parameter within the set of frequency domain signals; and

classifying atrial fibrillation based on the predetermined parameter.

23. The method as set forth in claim 22, further comprising receiving the electrical signals by placing at least one electrode on an external surface of a patient and detecting the electrical signals with the at least one electrode.

24. The method as set forth in claim 22, further comprising receiving the electrical signals by placing the at least one electrode within a patient and detecting the fibrillatory baseline signal.

25. The method as set forth in claim 22, further comprising receiving electrical signals by receiving digital signals indicative of the heart'"'"'s activity.

26. The method as set forth in claim 22, further comprising receiving electrical signals by receiving analog signals indicative of the heart'"'"'s electrical activity.

27. The method as set forth in claim 22, further comprising processing the electrical signals by detecting QRST complexes in the electrical signals, generating one or more QRST templates, and subtracting QRST complexes using the one or more QRST templates.

28. The method as set forth in claim 22, wherein the fibrillatory baseline signal comprises a digital signal, and further comprising processing the electrical signals by compressing the fibrillatory baseline signal.

29. The method as set forth in claim 28, further comprising processing the electrical signals by applying a Hamming window.

30. The method as set forth in claim 22, further comprising transforming the fibrillatory baseline signal by performing a Fourier transformation of the fibrillatory baseline signal to produce the set of frequency domain signals.

31. The method of claim 22, further comprising:

detecting a transition in power level within the set of frequency domain signals; and

classifying atrial fibrillation based upon the detected transition.

32. The method as set forth in claim 31, further comprising classifying atrial fibrillation by measuring a midpoint between mean values of high and low frequencies of the set of frequency domain signals.

33. The method as set forth in claim 31, further comprising classifying atrial fibrillation by measuring a band of frequencies having an abrupt fall in power levels over a predetermined frequency range.

34. The method as set forth in claim 33, further comprising classifying atrial fibrillation by measuring the abrupt fall in power levels over a frequency range of about 1 Hertz.

35. The system as et forth in claim 33, further comprising classifying atrial fibrillation by measuring abrupt falls in power levels characterized by decreases in power of over about one-half of an order of magnitude.

36. The method as set forth in claim 31, further comprising indicating results after classifying the atrial fibrillation.

37. The method as set forth in claim 31, further comprising classifying a patient in atrial fibrillation with a transition at less than a predetermined frequency as having a greater than fifty percent probability of converting to normal sinus rhythm upon an administration of an antiarrhythmic drug.

38. The method of claim 37, further comprising classifying a patient in atrial fibrillation with a transition at less than a predetermined frequency as having substantially a one-hundred percent probability of converting to normal sinus rhythm upon acute administration of ibutilide.

39. The method of claim 37, wherein the predetermined value is about six Hertz.

40. The method of claim 31, further comprising classifying a patient in atrial fibrillation with a transition at more than a predetermined value as having less than a fifty percent probability of converting to normal sinus rhythm upon an administration of an antiarrhythmic drug.

41. The method of claim 40, further comprising classifying a patient in atrial fibrillation with a transition at more than a predetermined value as having about a twenty-eight percent probability of converting to normal sinus rhythm upon an administration of ibutilide.

42. The method of claim 40, wherein the predetermined value is about six Hertz.

43. The method as set forth in claim 31, further comprising classifying a patient in atrial fibrillation with a transition at less than a predetermined frequency as having a greater than fifty percent probability of remaining in normal sinus rhythm for more than about three months after successful cardioversion, said cardioversion achieved by either electric shock or by an administration of an antiarrhythmic drug.

44. The method of claim 43, further comprising classifying a patient in atrial fibrillation with a transition at less than a predetermined frequency as having about an eighty percent probability of remaining in normal sinus rhythm for more than about three months after a successful cardioversion either by an electric shock or by an administration of an antiarrhythmic drug.

45. The method of claim 43, wherein the predetermined frequency is about six and one-half Hertz.

46. The method as set forth in claim 31, further comprising classifying a patient in atrial fibrillation with a transition at more than a predetermined frequency as having a greater than fifty percent probability of recurrent atrial fibrillation within about three months after a successful cardioversion, said cardioversion achieved by either an electric shock or by an administration of an antiarrhythmic drug.

47. The method of claim 46, further comprising classifying a patient in atrial fibrillation with a transition at more than a predetermined frequency as having about an eighty percent probability of recurrent atrial fibrillation within about three months after a successful cardioversion either by an electric shock or by an administration of an antiarrhythmic drug.

48. The method of claim 46, wherein the predetermined frequency is about six and one-half Hertz.

49. A computer-readable medium storing software for classifying atrial fibrillation, the software being adapted to receive electrical signals indicative of a heart'"'"'s activity, process the electrical signals to remove ventricular activity and to output a fibrillatory baseline signal, transform the fibrillatory baseline signal into a set of frequency domain signals, and detect a value of a predetermined parameter within the set of frequency domain signals.

50. The computer-readable medium of claim 49, the software being further adapted to detect a transition which is characterized by a change in power within the set of frequency domain signals.

51. The computer-readable medium as set forth in claim 50, the software being further adapted to classify a severity of the atrial fibrillation based on a frequency at which the transition occurs.

52. The computer-readable medium as set forth in claim 51, the software being further adapted to classify a patient in atrial fibrillation with a transition at less than a predetermined frequency as having more than a fifty percent probability of converting to normal sinus rhythm upon an administration of an antiarrhythmic drug.

53. The computer-readable medium of claim 52, the software being further adapted to classify a patient in atrial fibrillation with a transition at less than a predetermined frequency as having substantially one-hundred percent probability of converting to normal sinus rhythm upon an administration of ibutilide.

54. The computer-readable medium of claim 52, wherein the predetermined frequency is about six Hertz.

55. The computer-readable medium of claim 51, the software being further adapted to classify a patient in atrial fibrillation with a transition at greater than a predetermined value as having less than a fifty percent probability of converting to normal sinus rhythm upon an administration of an antiarrhythmic drug.

56. The computer-readable medium of claim 55, the software being further adapted to classify a patient in atrial fibrillation with a transition at greater than a predetemined value as having about a twenty eight percent probability of converting to normal sinus rhythm upon an administration of ibutilide.

57. The computer-readable medium of claim 55, wherein the predetermined frequency is about six Hertz.

58. The computer-readable medium as set forth in claim 51, the software being further adapted to classify a patient in atrial fibrillation with a transition at less than a predetermined frequency as having greater than a fifty percent probability of remaining in normal sinus rhythm for more than about three months after a successful cardioversion, said cardioversion achieved by either an electric shock or by an administration of an antiarrhythmic drug.

59. The computer-readable medium as set forth in claim 58, the software being further adapted to classify a patient in atrial fibrillation with a transition at less than a predetermined frequency as having about an eighty percent probability of remaining in normal sinus rhythm for more than about three months after a successful cardioversion by either an electric shock or by an administration of an antiarrhythmic drug.

60. The computer-readable medium of claim 58, wherein the predetermined frequency is about six and one-half Hertz.

61. The computer-readable medium as set forth in claim 51, the software being further adapted to classify a patient in atrial fibrillation with a transition at more than a predetermined frequency as having greater than a fifty percent probability of recurrent atrial fibrillation within about three months after a successful cardioversion, said cardioversion achieved by either an electric shock or by an administration of an antiarrhythmic drug.

62. The computer readable medium of claim 61, the software being further adapted to classify a patient in atrial fibrillation with a transition at more than a predetermined frequency as having about an eighty percent probability of recurrent atrial fibrillation within about three months after a successful cardioversion by either an electric shock or an administration of an antiarrhythmic drug.

63. The computer-readable medium of claim 61, wherein the predetermined frequency is about six and one-half Hertz.

64. The computer-readable medium as set forth in claim 51, the software being further adapted to classify atrial fibrillation by measuring a midpoint between mean values of high and low frequencies of the set of frequency domain signals.

65. The computer-readable medium as set forth in claim 51, the software being further adapted to classify atrial fibrillation by measuring a band of frequencies having an abrupt fall in power levels over a predetermined frequency range.

66. The computer-readable medium as set forth in claim 65, the software being further adapted to classify atrial fibrillation by measuring the abrupt fall in power levels over a frequency range of about 1 Hertz.

67. The computer-readable medium as set forth in claim 66, the software being further adapted to classify atrial fibrillation by measuring abrupt falls in power levels characterized by decreases in power of over about one-half of an order of magnitude.

68. The computer-readable medium as set forth in claim 51, the software being further adapted to indicate results after classifying the atrial fibrillation.