Hybrid Convolutional Neural Network for Localization of Epileptic Focus Based on iEEG

Table 2

Detection results of focal and nonfocal EEG signals of published journal articles using the Bern-Barcelona EEG database.

Author (year)	Feature extraction methods	Classifier	Accuracy
Sharma et al. (2015) [21]	EMD, entropy	SVM	87.0%
Sriraam et al. (2017) [22]	Statistical, frequency-based, entropy, FD, Wilcoxon test	SVM	92.2%
Sharma et al. (2017) [23]	WFB, entropy, -test	LS-SVM	94.3%
Das and Bhuiyan (2016) [24]	EMD-DWT, entropy	KNN	89.4%
Bhattacharyya et al. (2017) [25]	EWT, RPS, CTM	LS-SVM	90.0%
Gupta et al. (2017) [26]	FAWT, entropy, Kruskal-Wallis test	LS-SVM	94.4%
Zhao et al. (2018) [27]	Entropy	CNN	83.0%
Daoud and Bayoumi (2020) [28]	DCAE	MLP	93.2%
TFCNN	STFT	2d-CNN	91.9%
Mixed-CNN	1d convolution layer	2d-CNN	92.5%
TF-HybridNet	STFT, 1d convolution layer	2d-CNN	94.3%