2016/04/26 by Zahra Roshan Zamir, Zamir, Z. Roshan
Computer Science · Neuroscience · #Blind Source Separation Techniques #EEG and Brain-Computer Interfaces #FOS: Computer and information sciences #FOS: Mathematics #Machine Learning (cs.LG) #Neural Networks and Applications #Optimization and Control (math.OC)
paper · pdf · doi:10.48550/arxiv.1604.08500
openalex publication_date 2016/04/26 · openalex created_date 2022/10/02 · openalex updated_date 2026/07/28
An epileptic seizure is a transient event of abnormal excessive neuronal\ndischarge in the brain. This unwanted event can be obstructed by detection of\nelectrical changes in the brain that happen before the seizure takes place. The\nautomatic detection of seizures is necessary since the visual screening of EEG\nrecordings is a time consuming task and requires experts to improve the\ndiagnosis. Four linear least squares-based preprocessing models are proposed to\nextract key features of an EEG signal in order to detect seizures. The first\ntwo models are newly developed. The original signal (EEG) is approximated by a\nsinusoidal curve. Its amplitude is formed by a polynomial function and compared\nwith the pre developed spline function.Different statistical measures namely\nclassification accuracy, true positive and negative rates, false positive and\nnegative rates and precision are utilized to assess the performance of the\nproposed models. These metrics are derived from confusion matrices obtained\nfrom classifiers. Different classifiers are used over the original dataset and\nthe set of extracted features. The proposed models significantly reduce the\ndimension of the classification problem and the computational time while the\nclassification accuracy is improved in most cases. The first and third models\nare promising feature extraction methods. Logistic, LazyIB1, LazyIB5 and J48\nare the best classifiers. Their true positive and negative rates are 1 while\nfalse positive and negative rates are zero and the corresponding precision\nvalues are 1. Numerical results suggest that these models are robust and\nefficient for detecting epileptic seizure.\n