Name
Title | ||
|---|---|---|
Seizure prediction using long-term fragmented intracranial canine and human EEG recordings |
Abstract | ||
|---|---|---|
This paper presents a novel patient-specific algorithm for prediction of seizures in epileptic patients. Spectral power features, including relative spectral powers and spectral power ratios, and cross correlation coefficients between all pairs of electrodes, are extracted as two independent feature sets. Both feature sets are selected independently in a patient-specific manner by classification and regression tree (CART). Selected features are further processed by a second-order Kalman filter and then input independently to three different classifiers which include AdaBoost, radial basis function kernel support vector machine (RBF-SVM) and artificial neural network (ANN). The algorithm is tested on the intra-cranial EEG (iEEG) from the recent American Epilepsy Society Seizure Prediction Challenge database. Intracranial EEG was recorded from five dogs and two patients. These datasets have varying numbers of electrodes and are sampled at different sampling frequencies. It is shown that the spectral feature set achieves a mean AUC of 0.7538, 0.7739, and 0.7948 for AdaBoost, SVM, and ANN, respectively. The correlation coefficients feature set achieves a mean AUC of 0.6640, 0.7403, and 0.7875 for AdaBoost, SVM, and ANN, respectively. The combined best results which use patient-specific feature sets achieve a mean AUC of 0.7603, 0.8472, and 0.8884 for AdaBoost, SVM, and ANN, respectively. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.1109/ACSSC.2016.7869060 | 2016 50th Asilomar Conference on Signals, Systems and Computers |
Keywords | Field | DocType |
seizure prediction,long-term fragmented intracranial canine EEG recordings,long-term fragmented human EEG recordings,patient-specific algorithm,epileptic patients,relative spectral powers,spectral power ratios,cross correlation coefficients,independent feature sets,classification and regression tree,CART,feature selection,second-order Kalman filter,AdaBoost,radial basis function kernel support vector machine,RBF-SVM,artificial neural network,ANN,sampling frequencies,spectral feature set,mean AUC,patient-specific feature sets | Decision tree,AdaBoost,Radial basis function kernel,Pattern recognition,Computer science,Support vector machine,Kalman filter,Feature extraction,Speech recognition,Artificial intelligence,Artificial neural network,Electroencephalography | Conference |
ISSN | ISBN | Citations |
1058-6393 | 978-1-5386-3955-9 | 0 |
PageRank | References | Authors |
0.34 | 9 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Zisheng Zhang | 1 | 28 | 3.34 |
| keshab k parhi | 2 | 3235 | 369.07 |