Detection and Classification Methods for EEG Epileptic Seizures

Abstract

Epilepsy is the maximum common brain disease which has been spreading largely around the arena. It takes place because of excessive or synchronous strange activities in human brain. To degree the electric seizure pastime of mind many technology and techniques had been developed. Nearly 80% of people with epilepsy stay in low-and center-earnings countries. It is estimated that as much as 70% of people residing with epilepsy should stay seizure-free if well diagnosed and handled. The chance of premature demise in humans with epilepsy becomes three times more than for the overall population. Three quarters of people with epilepsy living in low-profits nations do now not get the remedy they want. In many parts of the arena, human beings with epilepsy be afflicted by stigma and discrimination. Epilepsy is characterized by means of surprising bursts of extra energy in the mind, manifesting as recurrent seizures. Epilepsy continues to be now not properly understood whilst compared with other neurological disorders. This paper provides an overview of detection and classification era for the reason of EEG seizure. Time frequency transforms and system learning plays an important function in extracting meaningful facts. The overall performance of detection and classification techniques of the researchers is the primary focus of the paper.

Segmentation and Classification of Brain Tumor from Magnetic Resonance Images Using K-Means Algorithm and Hybrid PSO-WCA Based Radial Basis Function Neural Network

“…Nilesh Bhaskarrao Bahadure et al [14] utilized support vector machine for classification and achieved 96.51% accuracy. The classification and detection of the brain tumors are presented by the researchers through dissimilar classifiers such as SVM [25,26],PNN, MLP [28] etc. and found classification results in terms of "accuracy" and "computational time" for the cancerous and noncancerous brain tumors.…”