Seizure warning and prediction

• US 6,304,775 B1
• Filed: 09/22/1999
• Issued: 10/16/2001
• Est. Priority Date: 09/22/1999
• Status: Expired due to Term

• Current Assignee: University of Florida Research Foundation, Incorporated (State University System of Florida)
• Original Assignee: James Chris Sackellares, Leonidas D. Iasemidis
• Inventors: Iasemidis, Leonidas D., Sackellares, James Chris
• Primary Examiner(s): Lacyk, John P.
• Assistant Examiner(s): SZMAL, BRIAN SCOTT
• Application Number:
  US09/400,982
• Time in Patent Office: 755 Days
• Field of Search: 600/544, 600/378, 600/300, 600/545, 607/45, 364/574
• US Class Current: 600/544
• CPC Class Codes: A61B 5/369 Electroencephalography [EEG] devices for psychotechnics A61B5/16
• CPC Class Codes: A61B 5/37 Intracranial electroencephalography [IC-EEG], e.g. electrocorticography [ECoG]
• CPC Class Codes: A61B 5/372 Analysis of electroencephalograms
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy

Abstract

Impending seizure warning (ISW), seizure susceptibility period detection (SSPD), hours or days before an impending seizure, and time to impending seizure prediction (TISP) are provided by measuring each of a number of signals from different locations about a patient'"'"'s brain, and generating therefrom, a spatio-temporal response based on these signals. Chaoticity profiles are then generated for each spatio-temporal response. Over a period of time, a determination is made as to whether a certain level of dynamic entrainment exists between the chaoticity profiles associated with the responses from a set of critical locations. If so, a seizure warning, an indication of seizure susceptibility and a prediction of a time of occurrence of an impending seizure may be issued.

Claims

1. A method of analyzing a multidimensional system comprising the steps of:

measuring each of a plurality of signals generated by said multidimensional system, wherein each of the plurality of signals represents a response associated with a corresponding spatial location within said multidimensional system;

generating a phase space representation for each of the plurality of signals;

deriving a signal profile for each of the plurality of signals, wherein each signal profile represents a level of chaoticity for each corresponding signal over time;

comparing each of the signal profiles;

selecting one or more groups of signals based on the comparison between their corresponding signal profiles; and

characterizing the state dynamics of the multidimensional system as a function of the signal profile comparisons associated with the selected one or more signal groups.

2. The method of claim 1, wherein said multidimensional system is the brain.

3. The method of claim 2, wherein said step of characterizing the state of the multidimensional system as a function of the signal profile comparisons associated with the selected one or more signal groups comprises the step of:

determining whether the brain is likely to experience a seizure.

4. The method of claim 3 further comprising the step of:

determining a time to impending seizure.

5. The method of claim 3 further comprising the step of:

deriving a time to impending seizure.

6. The method of claim 1 further comprising the step of:

issuing a warning of a state transition based on the characterized state dynamics of the multi-dimensional system.

7. The method of claim 1 further comprising the step of:

predicting a time of occurrence of a state transition based on the characterized state dynamics of the multi-dimensional system.

8. The method of claim 1, wherein said step of deriving a signal profile for each of the plurality of signals comprises the step of:

determining a plurality of Lyapunov exponent values over time for each of the plurality of signals.

9. The method of claim 1 further comprising the step of:

deriving additional signal profiles for each of the plurality of signals, wherein each of the signal profiles associated with a corresponding signal represents a level of chaoticity for that corresponding signal based on a different Lyapanov exponent.

10. The method of claim 9, wherein one of the signal profiles associated with each signal represents a level of chaoticity for the corresponding signal based on a maximum Lyapunov exponent.

11. The method of claim 9, wherein the number of signal profiles derived for a corresponding signal equals the number of dimensions of said multidimensional system.

12. The method of claim 1 wherein said step of comparing each of the signal profiles comprises the step of:

determining a degree of correlation between the signal profiles.

13. The method of claim 12, wherein said step of determining a degree of correlation between the signal profiles comprises the step of:

determining a level of entrainment between the signal profiles.

14. The method of claim 13, wherein said step of determining the level of entrainment between the signal profiles comprises the step of:

applying a T-index statistic to quantify the level of entrainment between the signal profiles.

15. The method of claim 13, wherein said step of determining the level of entrainment between the signal profiles comprises the step of:

applying an F-index statistic to quantify the level of entrainment between the signal profiles.

16. The method of claim 13, wherein said step of determining the level of entrainment between the signal profiles comprises the step of:

employing neural network to quantify the level of entrainment between the signal profiles.

17. The method of claim 13, wherein said step of determining the level of entrainment between the signal profiles comprises the step of:

employing pattern recognition techniques to quantify the level of entrainment between the signal profiles.

18. The method of claim 13, wherein said step of determining a level of entrainment between the signal profiles comprises the step of:

determining a level of phase entrainment between the signal profiles.

19. The method of claim 13, wherein said step of determining a level of entrainment between the signal profiles comprises the step of:

determining a level of entrainment between the signal profiles using higher order derivatives of the signal profiles.

20. A method for providing seizure warnings comprising the steps of:

acquiring a time series signal from each of a plurality of locations about the brain, wherein each signal and its corresponding location constitute a corresponding channel;

for each channel, generating a spatio-temporal response based on the corresponding time series signal;

quantifying a sequence of chaoticity values for each channel based on the corresponding spatio-temporal response, wherein each sequence of chaoticity values constitutes a chaoticity profile;

comparing, over time, the chaoticity profiles associated with each of a number of channel pairs;

evaluating, over time, levels of entrainment between the chaoticity profiles associated with each of the channel pairs;

determining whether the levels of entrainment associated with one or more of the channel pairs are statistically significant; and

generating a seizure warning if it is determined that the levels of entrainment associated with one or more of the channel pairs are statistically significant.

21. The method of claim 20 further comprising the step of:

selecting a number of critical channel pairs during an initialization period, wherein a critical channel pair is identified as a pair of channels whose corresponding chaoticity profiles exhibit a relatively high level of entrainment prior to a seizure; and

wherein each of said number of channel pairs is a critical channel pair.

22. The method of claim 21 further comprising the steps of:

refining the selection of critical channel pairs after a seizure; and

using the refined selection of critical channel pairs in generating a seizure warning for a next seizure.

23. The method of claim 20, wherein said step of acquiring a time series signal from each of the plurality of locations about the brain comprises the step of:

measuring electrical signals from each of the locations about the brain using electroencephalography.

24. The method of claim 20, wherein said step of acquiring a time series signal from each of the plurality of locations about the brain comprises the step of:

measuring electromagnetic signals from each of the locations about the brain using magneto-electroencephalography.

25. The method of claim 20 further comprising the step of:

digitizing each of the acquired time series signals.

26. The method of claim 20, wherein said step of generating a spatio-temporal response based on the corresponding time series signal comprises the step of:

generating a p-dimensional phase space portrait from each time series signal using the Method of Delays.

27. The method of claim 20, wherein said step of quantifying a sequence of chaoticity values for each channel based on the corresponding spatio-temporal response, wherein each sequence of chaoticity values constitutes a chaoticity profile, comprises the step of:

computing a sequence of Lyapunov exponents; and

generating a sequence of average Lyapunov exponents by averaging the Lyapunov exponents over “

sliding time windows”

.

28. The method of claim 27 further comprising the step of:

quantifying a plurality of chaoticity value sequences, each chaoticity value sequence constituting a distinct chaoticity profile, wherein a chaoticity value sequence is produced for each channel based on a corresponding spatio-temporal response, and wherein each of the chaoticity profiles associated with each channel is based on a different Lyapunov exponent.

29. The method of claim 28, wherein one chaoticity profile associated with each channel is based on a maximum Lyapunov exponent.

30. The method of claim 28, wherein the number of chaoticity profiles quantified for each channel equals the number of dimensions p being used to characterize the brain.

31. The method of claim 20, wherein said step of comparing, over time, the chaoticity profiles associated with each of the number of channel pairs comprises the step of:

generating a T-index profile for each of the number of channel pairs based on the chaoticity profiles associated with each channel pair.

32. The method of claim 31, wherein said step of evaluating, over time, the levels of entrainment between the chaoticity profiles associated with each of the channel pairs comprises the steps of:

comparing a sequence of T-index values associated with each T-index for each of the number of channel pairs to a T-index threshold value.

33. The method of claim 32, wherein said step of determining whether the levels of entrainment associated with one or more of the channel pairs are statistically significant comprises the step of:

determining whether the T-index values associated with each T-index for the one or more channel pairs is less than the T-index threshold value for a given amount of time, said given amount of time and T-index threshold being selected based on a desirable level of statistical significance.

34. The method of claim 20, wherein said step of generating a seizure warning, if it is determined that the levels of entrainment associated with one or more of the channel pairs are statistically significant, comprises the step of:

generating an impending seizure warning during a preictal stage of a next seizure.

35. The method of claim 20, wherein said step of generating a seizure warning, if it is determined that the levels of entrainment associated with one or more of the channel pairs are statistically significant, comprises the step of:

generating a seizure susceptibility period warning during an interictal stage.

36. The method of claim 20, wherein said step of generating a seizure warning, if it is determined that the levels of entrainment associated with one or more of the channel pairs are statistically significant, comprises the step of:

generating a time to seizure warning.

37. A method of activating a seizure interdiction device comprising the steps of:

acquiring each of a plurality of signals from a corresponding location of a patient'"'"'s brain, wherein each signal constitutes a separate channel;

for each channel, generating a spatio-temporal response based on the corresponding signal;

generating a chaoticity profile, comprising a sequence of chaoticity values, for each channel based on the corresponding spatio-temporal response;

determining whether a level of entrainment between chaoticity profiles associated with a critical channel pair is statistically significant;

generating a seizure warning if it is determined that the level of entrainment associated with the critical channel pair is statistically significant; and

triggering the seizure interdiction device to deliver an antiseizure treatment to the patient if a seizure warning is generated.

38. The method of claim 37, wherein said step of triggering the seizure interdiction device to deliver an antiseizure treatment to the patient if a seizure warning is generated comprises the step of:

delivering an electrical or electromagnetic stimulus to the patient'"'"'s brain, vagus nerve or other neural structure to abort an impending seizure.

39. The method of claim 37, wherein said step of triggering the seizure interdiction device to deliver an antiseizure treatment to the patient if a seizure warning is generated comprises the step of:

releasing into the patient a compound to abort an impending seizure.

40. The method of claim 37 further comprising the steps of:

comparing the chaoticity profiles associated with a number of channel pairs;

evaluating a level of entrainment between the chaoticity profiles associated with each of the number of channel pairs; and

selecting one or more critical channel pairs from the number of channel pairs, wherein a critical channel pair is one which exhibits a statistically significant level of entrainment, or a maximum level of entrainment with respect to other channel pairs, during a preictal stage of a seizure.

41. The method of claim 40 further comprising the step of:

repeating, after a seizure, the steps of evaluating a level of entrainment between the chaoticity profiles associated with each of the number of channel pairs and selecting one or more critical channel pairs from the number of channel pairs;

updating the one or more critical channel pairs; and

generating a seizure warning based on the level of entrainment associated with the updated one or more critical channel pairs for a next seizure.

42. The method of claim 37, wherein the sequence of chaoticity values comprise a sequence of Lyapunov exponent values.

43. The method of claim 37, wherein the seizure interdiction device is an implantable device.

44. An apparatus for providing seizure interdiction comprising:

a plurality of sensors coupled to a patient'"'"'s head, said sensors detecting signals from a corresponding location of the patient'"'"'s brain;

processing means for generating a seizure warning based on the plurality of signals detected by said plurality of sensors, said processing means comprising,

means for receiving the plurality of signals detected by said plurality of sensors,

means for preprocessing the plurality of signals detected by said sensors so as to produce a digital equivalent for each of said signals,

means for generating a spatio-temporal response for each of a corresponding one of the plurality of digital signals,

means for generating a chaoticity profile, comprising a sequence of chaoticity values, from each spatio-temporal response,

means for determining whether a level of entrainment between chaoticity profiles associated with a critical pair of signals is statistically significant;

means for generating a seizure warning if it is determined that the level of entrainment associated with the critical signal pair is statistically significant; and

a seizure interdiction device coupled to said processing means, said seizure interdiction device comprising means for delivering an antiseizure treatment to the patient if a seizure warning is generated.

45. The seizure interdiction apparatus of claim 44, wherein said seizure interdiction device is an implantable device.

46. The seizure interdiction apparatus of claim 45, wherein said implantable seizure interdiction device comprises:

means for delivering an electrical or electromagnetic stimulus to the patient'"'"'s brain, vagus nerve or other neural structure to abort an impending seizure, if a seizure warning is generated.

47. The seizure interdiction apparatus of claim 45, wherein said implantable seizure interdiction device comprises:

means for releasing into the patient a drug to abort an impending seizure, if a seizure warning is generated.