Method and system for predicting and preventing seizures

• US 20040267152A1
• Filed: 05/14/2004
• Published: 12/30/2004
• Est. Priority Date: 02/26/2003
• Status: Active Grant

• Current Assignee: Jaime A. Pineda
• Original Assignee: Jaime A. Pineda
• Inventors: Pineda, Jaime A.
• Granted Patent: US 7,269,455 B2
• Time in Patent Office: Days
• US Class Current: 600/544
• CPC Class Codes: A61B 5/374 Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy
• CPC Class Codes: A61B 5/7267 involving training the classification device

Abstract

A method and system for the detection and prevention of epileptic seizures utilizing bioelectric signals to assess a seizure profile (SP) and an adaptive control system for neurofeedback therapy. The inventive method and system provide the detection of changes in the non-linear dynamics of brain electrical activities to characterize and differentiate individual susceptibility to seizure onset, predict the occurrence of a seizure episode, and initiate neurofeedback training to prevent the attack.

Claims

1. A method for the detection of seizures comprising the steps of:

(i) acquiring a first bioelectric signal and a second bioelectric signal of a human subject;

(ii) processing the first of bioelectric signals to define a cognitive state of mind window for the subject;

(iii) processing the second bioelectric signal to define a moving cognitive state of mind window for the subject; and

(iv) comparing the baseline EEG window with the moving EEG window to define a dynamical similarity index indicative of the likelihood of a seizure.

2. The method according to claim 1 wherein the bioelectric signals comprises Electroencephalograph rhythms (EEG'"'"'s).

3. The method according to claim 2, wherein the EEG'"'"'s comprises at least one of mu, theta, alpha, N1. P3, SSVER signals.

4. The method according to claim 1 wherein the EEG'"'"'s are processed according to time and frequency domain subcomponents with at least of one the techniques selected from Independent Component Analysis, Variable Epocl Frequency Decomposition (VEFD), Fast Fourier Transformation, Event-Related Potentials, Time-Frequency Expansion, and Feature Coherence Analysis.

5. The method according to claim 1 wherein the dynamical similarity index comprises historical changes of the bioelectric signal in the moving EEG windows to define a seizure profile.

6. The method according to claim 1 wherein nonlinear dynamics analyses are applied in defining the dynamical similarity index.

7. The method according to claim 6 wherein nonlinear dynamics analyses are supported by a cross correlation integral in the form of:

8. The method according to claim 5 wherein the seizure profile is compared with a selectively predefined threshold to activate at least one of an audio or a visual signal.

9. The method according to claim 8 which further comprises the step of administering a therapeutic anti-epileptic drug or vagal nerve stimulation to the subject.

10. The method according to claim 5, which further comprises the step of mapping the seizure profile onto a set of microprocessor system commands for controlling the external environment of the subject.

11. The method according to claim 10 wherein the microprocessor system commands further activate at least one of an audio signal or a visual signal therapeutically communicated to the subject.

12. The method according to claim 1 wherein the dynamical similarity index analysis is optionally applied to neural network analysis to define the correlation between the bioelectric signals and the occurrence of pre-seizure states.

13. The method according to claim 12 wherein the correlation between the bioelectric signals and the occurrence of pre-seizure states are supported by a Neural Neworks (NN) function.

14. The method according to claim 13 wherein the NN function is in the form of

15. The method according to claim 12 wherein the correlation between the bioelectric signals and the occurrence of pre-seizure states are supported by at least one of the algorithmic functions selected from learning vector quantization, radial basis function, Hopfield, and Kohonen.

16. A system for the detection of brain electrical activity for the predicting and preventing seizures of a human subject, comprising:

(i) signal acquisition means for acquiring a first set of bioelectric signals and a second set of bioelectric signals of a human subject, and

(ii) a computer means in communication with the signal acquisition means for processing the bioelectric signals to define a baseline cognitive state of mind and a moving state of mind and to compare the baseline and the moving states of mind for determining a dynamical similarity index (DSI) indicative of the likelihood of a seizure of the subject.

17. The system according to claim 16 wherein the bioelectric signals comprising an electroencephalogram (EEG) rhythm.

18. The system according to claim 17 wherein the EEG rhythm comprises at least one of mu, theta, alpha, N1. P3, SSVER signals.

19. The system according to claim 16 wherein the microprocessor-based computer is adaptable to process the EEG'"'"'s according to time and frequency domain subcomponents using at least of one the techniques selected from Independent Component Analysis, Variable Epocl Frequency Decomposition (VEFD), Fast Fourier Transformation, Event-Related Potentials, Time-Frequency Expansion, and Feature Coherence Analysis.

20. The system according to claim 16 wherein the dynamical similarity index comprises historical changes of the baseline and moving cognitive states of mind to define a seizure profile.

21. The system according to claim 16 which the bioelectric acquisition means comprises at least one bioelectric sensor in communication with the microprocessor based computer for the acquisition and analysis of the bioelectric signals.

22. The system according to claim 21 wherein the at least one bioelectric sensor is selectably placed in the proximity of selected locations of the body of the subject for the acquisition of bioelectric signals.

23. The system according to claim 21 wherein the sensors-to-microprocessor communication is selectively effected by wired or wireless means.

24. The system according to claim 23 wherein the wireless means including radio-frequency or infrared data association with built-in Object Exchange Protocol.

25. The system according to claim 21, which further comprises at least an audio enunciating or a visual display means when the seizure profile is out of conformance with a selectively predefined threshold.

26. The system according to claim 21, which further comprises feedback means to the subject upon the onset of a seizure.

27. The system according to claim 21, which further comprises controlling means for selectively delivering at least one of anti-epileptic drug therapy or vagal nerve stimulation to the subject.

28. The system according to claim 21 wherein the computing means comprises a mapping algorithm for mapping a dynamical similarity index or a seizure profile onto a set of microprocessor system commands for controlling the external environment of the human subject.