Title :
A systematic approach to seizure prediction using genetic and classifier based feature selection
Author :
Alessandro, M.D. ; Vachtsevanos, G. ; Esteller, R. ; Echauz, J. ; Sewell, D. ; Litt, B.
Author_Institution :
Dept. of Electr. & Comput. Eng., Georgia Inst. of Technol., Atlanta, GA, USA
Abstract :
Currently, there is no standard approach for evaluating the intracranial encephalographic signals for seizure prediction. This study evaluates the IEEG signals by applying a systematic approach to feature selection, classification and validation to predict seizures. After preprocessing and processing, a genetic algorithm selects reasonable features off-line from a preselected group of features to serve as inputs to the classifier based feature selection process. A probabilistic neural network is used to select the optimal feature vector using a reed forward sequential approach on the training data followed by classification. A study of four patients resulted in a 62.5% average probability of prediction and a block false positive rate of 0.2775 false positive predictions per hour.
Keywords :
electroencephalography; feature extraction; feedforward neural nets; genetic algorithms; learning (artificial intelligence); medical signal processing; patient monitoring; probability; signal classification; IEEG signals; average prediction probability; block false positive rate; classifier based feature selection; false positive predictions; feature classification; feature selection; feature validation; feed forward sequential approach; genetic algorithm; genetic based feature selection; intracranial encephalographic signals; optimal feature vector; patients; probabilistic neural network; seizure prediction; signal preprocessing; signal processing; training data; Biological neural networks; Biomedical computing; Biomedical engineering; Data preprocessing; Electroencephalography; Epilepsy; Feedforward neural networks; Genetic algorithms; Hospitals; Neural networks;
Conference_Titel :
Digital Signal Processing, 2002. DSP 2002. 2002 14th International Conference on
Print_ISBN :
0-7803-7503-3
DOI :
10.1109/ICDSP.2002.1028162