Method and apparatus for detection of nervous system disorders

• US 7,761,145 B2
• Filed: 12/12/2006
• Issued: 07/20/2010
• Est. Priority Date: 04/21/2006
• Status: Active Grant

• Current Assignee: Medtronic Incorporated (Medtronic Plc)
• Original Assignee: Medtronic Incorporated (Medtronic Plc)
• Inventors: Virag, Nathalie, Vetter, Rolf
• Primary Examiner(s): Nasser; Robert L
• Assistant Examiner(s): D'ANGELO, MICHAEL J
• Application Number:
  US11/609,420
• Publication Number:
  US 20070249954A1
• Time in Patent Office: 1,316 Days
• Field of Search: 600544-545
• US Class Current: 600/544
• CPC Class Codes: A61B 5/02405 Determining heart rate variability
• CPC Class Codes: A61B 5/352 Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval
• CPC Class Codes: A61B 5/369 Electroencephalography [EEG] devices for psychotechnics A61B5/16
• CPC Class Codes: A61B 5/375 using biofeedback
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy
• CPC Class Codes: A61B 5/7221 Determining signal validity, reliability or quality A61B5/7207 and A61B5/7217 take precedence
• 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 1/36082 Cognitive or psychiatric applications, e.g. dementia or Alzheimer's disease
• CPC Class Codes: G16H 20/30 relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
• CPC Class Codes: G16H 40/67 for remote operation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Systems and methods for detecting and/or treating nervous system disorders, such as seizures, are disclosed. Certain embodiments of the invention relate generally to implantable medical devices (IMDs) adapted to detect and treat nervous system disorders in patients with an IMD. Certain embodiments of the invention include detection of seizures based upon comparisons of long-term and short-term representations of physiological signals. Other embodiments include prediction of seizure activity based upon analysis of physiological signal levels. An embodiment of the invention monitors the quality of physiological signals, and may be able to compensate for signals of low signal quality. A further embodiment of the invention includes detection of seizure activity following the delivery of therapy.

Claims

1. A method of predicting a seizure event, the method comprising:

acquiring an electroencephalograph (EEG) signal and extracting one or more EEG features;

acquiring a cardiovascular (CV) signal and extracting one or more CV features;

comparing at least one of the EEG and CV features to one or more features extracted from a reference seizure to form at least one similarity measure;

comparing at least one of the EEG and CV features to one or more features extracted from a reference baseline to form at least one dissimilarity measure;

deriving a discriminant measure, the discriminant measure being a function of the at least one similarity measure and the at least one dissimilarity measure;

comparing the discriminant measure to a predetermined threshold; and

predicting a seizure event when the discriminant measure exceeds the predetermined threshold.

2. The method of claim 1 wherein deriving the discriminant measure includes determining a weighted sum of the EEG features and CV features extracted.

3. The method of claim 1 wherein the EEG features extracted include information determined from zero crossings of the EEG signal.

4. The method of claim 3 wherein the EEG features extracted include a measure of time intervals between zero crossings of the EEG signals.

5. The method of claim 4 wherein the EEG features extracted include a statistical representation of the time intervals between zero crossings over a time frame.

6. The method of claim 5 wherein the EEG features extracted include a mean value of the time intervals between zero crossings in the same direction.

7. The method of claim 5 wherein the EEG features extracted include a standard deviation of the time intervals between zero crossings in the same direction.

8. The method of claim 1 wherein the EEG features are extracted from one or more frequency sub-bands of the EEG signal.

9. The method of claim 8 wherein at least one of the frequency sub-bands is within a range of frequencies from about 0-50 Hz.

10. The method of claim 8 wherein at least one of the frequency sub-bands comprises a range of frequencies from about 12-40 Hz.

11. The method of claim 8 wherein the EEG features extracted include information determined from zero crossings of the EEG signal.

12. The method of claim 8 wherein the EEG features extracted from one or more frequency sub-bands are compared to each other to compute a measure of synchronization for use in deriving the discriminant measure.

13. The method of claim 12 wherein the measure of synchronization is computed using a cross-correlation function.

14. The method of claim 1 wherein a seizure event is predicted when the discriminant measure exceeds a predetermined threshold for a predetermined duration.

15. The method of claim 1 wherein the reference seizure is capable of being updated with information from a recent seizure event.

16. The method of claim 1 wherein the reference seizure is derived from one or more previous seizure events.

17. The method of claim 1 wherein the similarity measure is a weighted sum of features extracted from the reference seizure, and wherein the reference seizure is derived from one or more previous seizure events in a particular patient.

18. The method of claim 1 wherein the reference baseline is capable of being updated with information from periods without seizure events.

19. The method of claim 1 wherein the dissimilarity measure is a weighted sum of features extracted from the reference baseline, and wherein the reference baseline is derived from one or more previous periods without seizure events in a particular patient.

20. The method of claim 1 wherein the CV signals comprise ECG signals.

21. The method of claim 20 wherein the CV features extracted include data derived from R-R intervals of the ECG signals.

22. The method of claim 21 wherein the data derived from R-R intervals include a mean heart rate.

23. The method of claim 21 wherein the data derived from R-R intervals include a measure of heart rate variability.

24. The method of claim 21 wherein the data determined from R-R intervals include a measure of heart rate trend.

25. The method of claim 20 wherein the CV features extracted include information describing the marginality of a cardiovascular signal.

26. The method of claim 25 wherein the marginality information includes statistical information about R-R intervals over predetermined time intervals.

27. The method of claim 25 wherein the marginality information includes the number of ectopic events over a predetermined time interval.

28. The method of claim 20 wherein the CV features extracted include an index of autonomic nervous system (ANS) modulation.

29. The method of claim 28 wherein the index of ANS modulation is determined by blind source separation.

30. The method of claim 1 wherein the CV signals comprise hemodynamic signals.

31. The method of claim 30 wherein the CV features extracted are derived from blood pressure measurements.

32. The method of claim 31 the CV features extracted include a measure of blood pressure trend.

33. The method of claim 1 further comprising acquiring at least two EEG signals, wherein the extracted features from the at least two EEG signal channels are compared to each other to compute a measure of synchronization for use in deriving the discriminant measure.