INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
Performance Analysis of DWT-Based Epileptic Seizure Detection
Authors Name:
Spoorthy.G
, Jalaja.S
Unique Id:
IJSDR2209148
Published In:
Volume 7 Issue 9, September-2022
Abstract:
Epilepsy is a central nervous system disorder that is well defined by the startling and atypical behavior of seizures and causes loss of consciousness. The Electroencephalogram (EEG) signal is particularly good for assessing neurogenic activity and is often used in central nervous system interfaces in the human brain, and neuron disease diagnosis. Among the three modules, discrete wavelet transforms(DWT), feature extraction (FE), and the classification of the support vector machine(SVM), the FE unit matches the relevant neurobiological areas of the EEG data using the 9/7th DWT and extracts the required features. Feature extraction classifies the extracted EEG information into four basic sub-bands namely, alpha, beta, gamma, and theta. Feature Extraction utilizes signal analysis methods and computer technologies to extract information from electroencephalography signals. The proposed design consists of medically converted EEG data, DWT, wavelet decomposition, higher-order statistics, a feature extraction (FE) module, and an SVM module unit. Among the various machine-learning techniques, the support vector machine differentiates between healthy and unhealthy illnesses such as epileptic seizures, SVM is deployed due to its high reliability & adaptation to the presence of arbitrary nonlinear decision limits. The posited designed system provides higher accuracy and reliability compared to the conventional methods designed before in time.
"Performance Analysis of DWT-Based Epileptic Seizure Detection", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.7, Issue 9, page no.926 - 932, September-2022, Available :http://www.ijsdr.org/papers/IJSDR2209148.pdf
Downloads:
000336257
Publication Details:
Published Paper ID: IJSDR2209148
Registration ID:201889
Published In: Volume 7 Issue 9, September-2022
DOI (Digital Object Identifier):
Page No: 926 - 932
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn