Method for the real-time identification of seizures in an electroencephalogram (EEG) signal

• US 10,433,752 B2
• Filed: 04/07/2010
• Issued: 10/08/2019
• Est. Priority Date: 04/07/2009
• Status: Active Grant

• Current Assignee: National University of Ireland Galway
• Original Assignee: National University of Ireland Galway
• Inventors: Faul, Stephen Daniel, Temko, Andriy, Marnane, William Peter, Lightbody, Gordon, Boylan, Geraldine Bernadette
• Primary Examiner(s): Berhanu, Etsub D
• Application Number:
  US13/262,891
• Publication Number:
  US 20120101401A1
• Time in Patent Office: 3,471 Days
• Field of Search: 600544, 600545
• 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 5/7267 involving training the classification device
• CPC Class Codes: G06F 2218/12 Classification; Matching
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

The present invention relates to a method for the real-time identification of seizures in an Electroencephalogram (EEG) signal. The method provides for patient-independent seizure identification by use of a multi-patient trained generic Support Vector Machine (SVM) classifier. The SVM classifier is operates on a large feature vector combining features from a wide variety of signal processing and analysis techniques. The method operates sufficiently accurately to be suitable for use in a clinical environment. The method may also be combined with additional classifiers, such a Gaussian Mixture Model (GMM) classifier, for improved robustness, and one or more dynamic classifiers such as an SVM using sequential kernels for improved temporal analysis of the EEG signal.

Claims

1. A method for the real-time identification of seizures in an Electroencephalogram (EEG) signal, the steps of the method comprising:

(a) receiving an EEG signal comprising a plurality of channels of EEG data;

(b) for each of the plurality of channels of EEG data, segmenting the data into sequential epochs, each of the sequential epochs having an overlap with its neighboring sequential epochs;

and for an initial epoch of each of the plurality of channels performing the following steps comprising;

(c) extracting forty five or more features from each of the plurality of channels of EEG data;

(d) generating a feature vector from the extracted features;

(e) passing the feature vector for each of the plurality of channels of EEG data separately through a multi-patient trained generic Support Vector Machine (SVM) classifier and generating SVM channel seizure outputs for each feature vector, in which the multi-patient trained generic support vector machine classifier is trained on EEG data representing all seizure types, over all channels and over all patient types;

(f) fusing the SVM channel seizure outputs for all channels thereby generating an SVM epoch seizure output, the SVM epoch seizure output indicative of a seizure activity present in that epoch across all channels; and

(g) repeating steps (c) to (f) for each of the subsequent sequential epochs thereby generating a sequence of SVM channel seizure outputs and SVM epoch seizure outputs; and

(h) providing an SVM epoch decision to a user, the SVM epoch decision being indicative of whether the EEG data indicates the occurrence of a seizure or not.

2. The method as claimed in claim 1 further comprising the additional steps of:

generating a reduced feature vector;

passing the reduced feature vector through a multi-patient trained Gaussian Mixture Model (GMM) classifier; and

generating a GMM channel seizure output.

3. The method as claimed in claim 2 further comprising the additional step of combining the SVM channel seizure output and GMM channel seizure output for each of the sequential epochs thereby generating a combined channel seizure output for each of the sequential epochs for each of the plurality of channels.

4. The method as claimed in claim 3 further comprising the additional step of fusing the combined channel seizure outputs for each of the sequential epochs across all channels thereby generating a combined epoch seizure output for each of the sequential epochs.

5. The method as claimed in claim 1 further comprising the additional step of converting each of the sequence of SVM channel seizure outputs into a binary channel seizure decision.

6. The method as claimed in claim 1 further comprising the additional step of converting each of the sequence of SVM channel seizure outputs to a normalised probabilistic value between 0 and 1.

7. The method as claimed in claim 1 in which the step of fusing the SVM channel seizure outputs comprises using a logical OR operator.

8. The method as claimed in claim 1 further comprising the additional step of applying a moving average filter to the sequence of SVM epoch seizure outputs.

9. The method as claimed in claim 1 further comprising the additional step of applying a collar technique to the sequence of SVM epoch seizure outputs.

10. The method as claimed in claim 1 in which the step of extracting features from the plurality of channels of EEG data comprises extracting features using speech recognition analysis techniques.

11. The method as claimed in claim 1 in which the step of extracting features from the plurality of channels of EEG data comprises extracting features using time domain analysis techniques, frequency domain analysis techniques and information theory analysis techniques.

12. The method as claimed in claim 1 in which the method comprises the additional step of passing the feature vector through a Support Vector Machine (SVM) with sequential kernels and generating a dynamic SVM channel seizure output.

13. The method as claimed in claim 1 further comprising passing the feature vector through the multi-patient trained generic SVM classifier and generating an SVM channel artefact output.

14. A computer program product stored on a non-transitory carrier having computer program instructions for causing a computing device to perform the method of claim 1.

15. An apparatus for the real-time identification of seizures in an Electroencephalogram (EEG) signal, the apparatus comprising:

means for receiving an EEG signal comprising a plurality of channels of EEG data;

means for segmenting the data of each channel into a plurality of sequential epochs, each of the plurality of sequential epochs having an overlap with its neighboring plurality of sequential epochs;

means for extracting forty five or more features from each of the plurality of channels of EEG data for each of the plurality of sequential epochs;

means for generating a feature vector from the extracted features for each of the plurality of sequential epochs for each of the channels of EEG data;

a multi-patient trained generic Support Vector Machine (SVM) classifier adapted to process the feature vector so as to generate SVM channel seizure outputs for each of the plurality of sequential epochs, wherein the multi-patient trained generic SVM classifier is trained on EEG data representing all seizure types, over all channels and over all patient types; and

means for fusing the SVM channel seizure outputs for each of the channels thereby generating an SVM epoch seizure output indicative of a seizure activity present in that epoch across all channels; and

means for providing an SVM epoch decision to a user, the SVM epoch decision being indicative of whether the EEG data indicates the occurrence of a seizure or not.

16. The apparatus as claimed in claim 15 further comprising:

means for generating a reduced feature vector;

means for passing the reduced feature vector through a multi-patient trained Gaussian Mixture Model (GMM) classifier and generating a GMM channel seizure output for each of the plurality of sequential epochs;

means for combining the SVM channel seizure output and GMM channel seizure output for each of the plurality of sequential epochs thereby generating a combined channel seizure output; and

means for fusing the combined channel seizure outputs for each of the plurality of sequential epochs across all of the plurality of channels of EEG data thereby generating a combined epoch seizure output for each of the plurality of sequential epochs.

17. The apparatus as claimed in claim 15 further comprising means to convert each of the SVM channel seizure outputs into a binary channel seizure decision.

18. The apparatus as claimed in claim 15 further comprising means to apply a moving average filter to a sequence of SVM epoch seizure outputs.

19. The apparatus as claimed in claim 15 further comprising means to apply a collar technique to a sequence of SVM epoch seizure outputs.