An Accurate Automatic Epileptic Seizure Diagnosis With Logistic Regression Using Electro encephalography Signals
|
|
Author:
|
PRITI BHAGAT, K.S.RAMESH, S T PATIL
|
Abstract:
|
The electroencephalogram (EEG) signals have been used for various neurological assessment applications, and these are helpful for epileptic diagnosis patients, such as deficiencies, disorders, and diseases associated with hominid brains. In this work, epileptic seizure patient’s neurological conditions are analyzed for real and accurate prior diagnosis. The EEG signal contains electrical activities of the human mind and the corresponding activity of body parts regardingthe nervous system. This EEG data are collected from signals-type, which are already recorded and loaded to datasets;a large number of datasets are difficult to interpolate with present implemented methods. In this investigation,certain epileptic seizure activities and reliable diagnosis have been proposed to confiscate the urgency of treatment. AnElastic Net Regression (ENR) is a trained model for classification,and Wavelet Deep Stacked (WDS) autoencoder act like a preprocessor. This is a fully supervised epileptic seizure forecasting method, achieves outstanding results like accuracy is improves by 98.78, specificity 92.32, and sensitivity 98.52. The result shows that the ENR- WDS model is the best classifier and gives excellent performance metrics compared to various classifiers. For betterdiagnosis and treatment, our study is beneficial for epileptic patients at abnormal conditions.
|
Keyword:
|
EEG signals, Epileptic seizures, ENR machine learning, WDS.
|
EOI:
|
-
|
DOI:
|
https://doi.org/10.31838/ijpr/2020.12.03.142
|
Download:
|
Request For Article
|
|
|