Method and system for detecting seizures using electroencephalograms

• US 6,735,467 B2
• Filed: 04/15/2002
• Issued: 05/11/2004
• Est. Priority Date: 04/15/2002
• Status: Expired due to Fees

• Current Assignee: Persyst Development Corp.
• Original Assignee: Persyst Development Corp.
• Inventors: Wilson, Scott B.
• Primary Examiner(s): WINAKUR, ERIC FRANK
• Application Number:
  US10/123,834
• Publication Number:
  US 20030195429A1
• Time in Patent Office: 757 Days
• Field of Search: 600/544, 128/920, 128/925, 702/19
• US Class Current: 600/544
• CPC Class Codes: A61B 5/369 Electroencephalography [EEG] devices for psychotechnics A61B5/16
• CPC Class Codes: A61B 5/372 Analysis of electroencephalograms
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy
• 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
• CPC Class Codes: Y10S 128/925 Neural network

Abstract

A method and a system are provided for detection of seizures by applying advanced numerical analysis techniques to digitized waveforms of an electroencephalogram (EEG) recording. In an embodiment, the advanced numerical analysis techniques implemented in the method and system for detecting seizures include a combination of matching pursuit, neural network rules, and connected-object clustering algorithms.

Claims

1. A method of detecting a seizure, comprising the steps of:

a) dividing a digitized waveform of an electroencephalogram (EEG) recording into a plurality of epochs each having a first predetermined duration;

(b) computing matching pursuit for a given one of the epochs to obtain a plurality of seizure atoms;

(c) describing the seizure atoms and the given epoch with at least one neural network (NN) rule;

(d) applying connected-object clustering to the epochs in a sliding window of a second predetermined duration to obtain a clustering result; and

(e) establishing a seizure point from the clustering result.

2. The method of claim 1, further comprising the step of determining whether the seizure point is valid.

3. The method of claim 1, further comprising the steps of repeating steps (b)-(e) and determining whether a succeeding seizure point is better than a preceding seizure point.

4. The method of claim 1, wherein the first predetermined duration is less than the second predetermined duration.

5. The method of claim 4, wherein the first predetermined duration is about 2 seconds.

6. The method of claim 4, wherein the second predetermined duration is about 60 seconds.

7. The method of claim 1, further comprising the step of setting an initial time T=0 prior to the step of computing matching pursuit for the given epoch.

8. The method of claim 7, further comprising the steps of determining whether unprocessed EEG data is available at the time T, and causing a time delay of a third predetermined duration if unprocessed EEG data is unavailable at the time T.

9. The method of claim 8, wherein the digital waveform is one of a plurality of waveforms forming a montage of the EEG recording, the method further comprising the steps of obtaining a subsequent waveform of the EEG recording and repeating steps (b)-(e) if unprocessed EEG data is available at the time T.

10. The method of claim 9, further comprising the step of establishing a seizure event with a plurality of proximal seizure points from the waveforms of the montage.

11. The method of claim 10, further comprising the step of determining whether the seizure event is valid.

12. The method of claim 11, further comprising the step of notifying a user that a seizure is identified if the seizure event is valid.

13. The method of claim 12, further comprising the step of saving the time T and a description of the seizure.

14. The method of claim 11, further comprising the steps of adding a time of the third predetermined duration to the time T and shifting the sliding window by the time of the third predetermined duration until the EEG recording is finished.

15. The method of claim 14, wherein the third predetermined duration is about 1 second.

16. A computer readable medium comprising a plurality of instructions, which when executed by a computer, cause the computer to perform the steps of:

(a) dividing a digitized waveform of an electroencephalogram (EEG) recording into a plurality of epochs each having a first predetermined duration;

(b) computing matching pursuit for a given one of the epochs to obtain a plurality of seizure atoms;

(c) describing the seizure atoms and the given epoch with at least one neural network (NN) rule;

(d) applying connected-object clustering to the epochs in a sliding window of a second predetermined duration to obtain a clustering result; and

(e) establishing a seizure point from the clustering result.

17. The computer readable medium as set forth in claim 16, further comprising instructions which cause the computer to perform the step of determining whether the seizure point is valid.

18. The computer readable medium as set forth in claim 16, further comprising instructions which cause the computer to perform the steps of:

repeating steps (b)-(e); and

determining whether a succeeding seizure point is better than a preceding seizure point.

19. The computer readable medium as set forth in claim 16, wherein the first predetermined duration is less than the second predetermined duration.

20. The computer readable medium as set forth in claim 19, wherein the first predetermined duration is about 2 seconds.

21. The computer readable medium as set forth in claim 19, wherein the second predetermined duration is about 60 seconds.

22. The computer readable medium as set forth in claim 16, further comprising instructions which cause the computer to perform the step of setting an initial time T=0 prior to the step of computing matching pursuit for the given epoch.

23. The computer readable medium as set forth in claim 22, further comprising instructions which cause the computer to perform the steps of:

determining whether unprocessed EEG data is available at the time T; and

causing a time delay of a third predetermined duration if unprocessed EEG data is unavailable at the time T.

24. The computer readable medium as set forth in claim 23, wherein the digitized waveform is one of a plurality of waveforms of the EEG recording which includes a montage of multiple channels of simultaneous waveforms, the computer readable medium further comprising instructions which cause the computer to perform the steps of:

obtaining a subsequent waveform of the EEG recording; and

repeating steps (b)-(e) if unprocessed EEG data is available at the time T.

25. The computer readable medium as set forth in claim 24, further comprising instructions which cause the computer to perform the step of establishing a seizure event with a plurality of proximal seizure points from the waveforms of the montage.

26. The computer readable medium as set forth in claim 25, further comprising instructions which cause the computer to perform the step of determining whether the seizure event is valid.

27. The computer readable medium as set forth in claim 26, further comprising instructions which cause the computer to perform the step of notifying a user that a seizure is identified if the seizure event is valid.

28. The computer readable medium as set forth in claim 27, further comprising instructions which cause the computer to perform the step of saving the time T and a description of the seizure.

29. The computer readable medium as set forth in claim 26, further comprising instructions which cause the computer to perform the steps of:

adding a time of the third predetermined duration to the time T; and

shifting the sliding window by the time of the third predetermined duration until the EEG recording is finished.

30. The computer readable medium as set forth in claim 29, wherein the third predetermined duration is about 1 second.

31. An article of manufacture, including a computer readable medium having computer readable program code means embodied therein for detecting a seizure, the computer readable program code means in the article of manufacture comprising:

(a) computer readable program code means for dividing a digitized waveform of an electroencephalogram (EEG) recording into a plurality of epochs each having a first predetermined duration;

(b) computer readable program code means for computing matching pursuit for a given one of the epochs to obtain a plurality of seizure atoms;

(c) computer readable program code means for describing the seizure atoms and the given epoch with at least one neural network (NN) rule;

(d) computer readable program code means for applying connected-object clustering to the epochs in a sliding window of a second predetermined duration to obtain a clustering result; and

(e) computer readable program code means for establishing a seizure point from the clustering result.

32. The article of manufacture as set forth in claim 31, wherein the computer readable program code means further comprises computer readable program code means for determining whether the seizure point is valid.

33. The article of manufacture as set forth in claim 31, wherein the computer readable program code means further comprises:

computer readable program code means for repeating steps performed by computer readable program code means (b)-(e); and

computer readable program code means for determining whether a succeeding seizure point is better than a preceding seizure point.

34. The article of manufacture as set forth in claim 31, wherein the first predetermined duration is less than the second predetermined duration.

35. The article of manufacture as set forth in claim 34, wherein the first predetermined duration is about 2 seconds.

36. The article of manufacture as set forth in claim 34, wherein the second predetermined duration is about 60 seconds.

37. The article of manufacture as set forth in claim 31, wherein the computer readable program code means further comprises computer readable program code means for setting an initial time T=0 prior to computing matching pursuit for the given epoch.

38. The article of manufacture as set forth in claim 37, wherein the computer readable program code means further comprises:

computer readable program code means for determining whether unprocessed EEG data is available at the time T; and

computer readable program code means for causing a time delay of a third predetermined duration if unprocessed EEG data is unavailable at the time T.

39. The article of manufacture as set forth in claim 38, wherein the digitized waveform is one of a plurality of waveforms of the EEG recording which includes a montage of multiple channels of simultaneous waveforms, and wherein the computer readable program code means further comprises:

computer readable program code means for obtaining a subsequent waveform of the EEG recording; and

computer readable program code means for repeating steps performed by computer readable program code means (b)-(e) if unprocessed EEG data is available at the time T.

40. The article of manufacture as set forth in claim 39, wherein the computer readable program code means further comprises computer readable program code means for establishing a seizure event with a plurality of proximal seizure points from the waveforms of the montage.

41. The article of manufacture as set forth in claim 40, wherein the computer readable program code means further comprises computer readable program code means for determining whether the seizure event is valid.

42. The article of manufacture as set forth in claim 41, wherein the computer readable program code means further comprises computer readable program code means for notifying a user that a seizure is identified if the seizure event is valid.

43. The article of manufacture as set forth in claim 42, wherein the computer readable program code means further comprises computer readable program code means for saving the time T and a description of the seizure.

44. The article of manufacture as set forth in claim 41, wherein the computer readable program code means further comprises:

computer readable program code means for adding the time of the third predetermined duration to the time T; and

computer readable program code means for shifting the sliding window by the time of the third predetermined duration until the EEG recording is finished.

45. The article of manufacture as set forth in claim 44, wherein the third predetermined duration is about 1 second.