Epilepsy Detection using Spectogram data and Convolutional Neural Networks

A neurological condition called epilepsy is characterised by aberrant brain activity, which can result in seizures or bursts of strange behaviours, feelings, or even unconsciousness. Epilepsy can strike any person. Men and women of different ages, ethnicities, and cultural backgrounds can develop epilepsy. The multi-channel electroencephalogram is used to look for certain patterns that can be used to identify epileptic activity, which is often done by a human specialist. Numerous attempts are made to automate this laborious and time-consuming operation using both traditional and Deep Learning techniques.After preprocessing the CHB-MIT dataset and selecting 5 patients at random, CNN model development is carried out.The model’s total average sensitivity for predicting epileptic seizures was 92.28% and 9.91% on average are false positives.This shows that the model accurately and successfully identified seizures in EEG readings.

Paper

Full text

PDF

Epilepsy Detection using Spectogram data and Convolutional Neural Networks

Semantic Scholar · Medicine · 2023

Abstract

A neurological condition called epilepsy is characterised by aberrant brain activity, which can result in seizures or bursts of strange behaviours, feelings, or even unconsciousness. Epilepsy can strike any person. Men and women of different ages, ethnicities, and cultural backgrounds can develop epilepsy. The multi-channel electroencephalogram is used to look for certain patterns that can be used to identify epileptic activity, which is often done by a human specialist. Numerous attempts are made to automate this laborious and time-consuming operation using both traditional and Deep Learning techniques.After preprocessing the CHB-MIT dataset and selecting 5 patients at random, CNN model development is carried out.The model’s total average sensitivity for predicting epileptic seizures was 92.28% and 9.91% on average are false positives.This shows that the model accurately and successfully identified seizures in EEG readings.

Similar papers

© 2026 NYSGPT2525 LLC