Epileptic EEG Classification by Using Time-Frequency Images for Deep Learning

Ozdemir, M. A. et al. (2021). Epileptic EEG Classification by Using Time-Frequency Images for Deep Learning, International Journal of Neural Systems.

Abstract

Epilepsy is one of the most common brain disorders worldwide. The most frequently used clinical tool to detect epileptic events and monitor epilepsy patients is the EEG recordings. There have been proposed many computer-aided diagnosis systems using EEG signals for the detection and prediction of seizures. In this study, a novel method based on Fourier-based Synchrosqueezing Transform (SST), which is a high-resolution time-frequency (TF) representation, and Convolutional Neural Network (CNN) is proposed to detect and predict seizure segments. SST is based on the reassignment of signal components in the TF plane which provides highly localized TF energy distributions. Epileptic seizures cause sudden energy discharges which are well represented in the TF plane by using the SST method. The proposed SST-based CNN method is evaluated using the IKCU dataset we collected, and the publicly available CHB-MIT dataset. Experimental results demonstrate that the proposed approach yields high average segment-based seizure detection precision and accuracy rates for both datasets (IKCU: 98.99% PRE and 99.06% ACC; CHB-MIT: 99.81% PRE and 99.63% ACC). Additionally, SST-based CNN approach provides significantly higher segment-based seizure prediction performance with 98.54% PRE and 97.92% ACC than similar approaches presented in the literature using the CHB-MIT dataset.

Keywords: Convolutional Neural Network (CNN), Deep Learning (DL), seizure detection, seizure prediction, segment-based, Synchrosqueezed Transform (SST), time-frequency images.

Content

This repository includes trained models of this work.

Test accuracy score: 99.21% with two clasess and input sizes of 128x128: Model_31

LATEST! Test accuracy score: 99.28% with two clasess and input sizes of 128x128: Model_32

Load The Model Weights in your project

Tutorial: Click

from keras.models import load_model

model = load_model('m32.h5')
model.summary()

Example Model inputs

DOI

https://doi.org/10.1142/S012906572150026X

Citation

Citation is now available. Please cite us by following;

@article{ozdemir2021epilepticEEG,
author = {Ozdemir, Mehmet Akif and Cura, Ozlem Karabiber and Akan, Aydin},
title = {Epileptic EEG Classification by Using Time-Frequency Images for Deep Learning},
journal = {International Journal of Neural Systems},
volume = {31},
number = {08},
pages = {2150026},
year = {2021},
doi = {10.1142/S012906572150026X},
note ={PMID: 34039254},
URL = {https://doi.org/10.1142/S012906572150026X},
publisher={World Scientific}

Cite from World Scientific or Cite from Google Scholar

Contact

If you need any help, feel free to start an issue (preferred because other people can benefit) or send me an email: [email protected]