Machine learning based classification of EEG signal for detection of child epileptic seizure without snipping

The electroencephalogram (EEG) signal is very important in the diagnosis of epilepsy. Long-term EEG recordings of an epileptic patient contain a huge amount of EEG data. Therefore, detecting epileptic activity is a very demanding process that requires a detailed analysis of the entire length of the...

Full description

Saved in:
Bibliographic Details
Published in:International journal of speech technology 2023-09, Vol.26 (3), p.559-570
Main Authors: Sethy, Prabira Kumar, Panigrahi, Millee, Vijayakumar, K., Behera, Santi Kumari
Format: Article
Language:English
Subjects:
Artificial Intelligence
Children
Classifiers
Electroencephalography
Engineering
Epilepsy
Machine learning
Performance evaluation
Seizures
Signal classification
Signal,Image and Speech Processing
Social Sciences
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The electroencephalogram (EEG) signal is very important in the diagnosis of epilepsy. Long-term EEG recordings of an epileptic patient contain a huge amount of EEG data. Therefore, detecting epileptic activity is a very demanding process that requires a detailed analysis of the entire length of the EEG data, usually performed by an expert. This paper evaluates machine learning classifiers' performance with their paradigms for classification of raw EEG signals into two classes, i.e., seizure and non-seizure. Here, the 13 descriptive features are taken into consideration and fed to the classifiers. Here, CHB-MIT Scalp EEG Database is used, which comprises paediatric subjects of 24 records. The performance of classifiers is evaluated categorically concerning gender and in total. The results confirmed that the fine KNN is the best classifier in males, females, and all subjects.
ISSN:1381-2416
1572-8110
DOI:10.1007/s10772-021-09855-7