1. Localization of epileptic foci from ieeg via mixed convolutional neural network
Authors : Linfeng Sui, Xuyang Zhao, Jianting Cao, Qibin Zhao
Pages : 8-13
DOI : http://dx.doi.org/10.21172/1.144.03
Keywords : EpilepsyFocus localizationCNNiEEG Abstract :Epileptic focus localization is plays a key role for successful surgical therapy of resection of epileptogenic tissues. However, manual diagnosis of intracranial electroencephalogram (iEEG) signals by highly skilled clinicians are arduous and time-consuming. In contrast, the focus can be localized by classifying of focal and non-focal iEEG signals, which can improve the accuracy and shorten the time to diagnosis. In this paper, we propose a mixed 1-D & 2-D convolutional neural networks (CNN) model which is inspired by recent developments from the field of image classification and attempt to improve the classification accuracy of iEEG signals. We apply our approach to the Bern- Barcelona iEEG dataset for evaluating the performance. Our model directly takes time-series iEEG as input and classifies the iEEG signals without requiring any feature extraction. Experimental results show that our approach is able to differentiate the focal from non- focal iEEG signals with an average classification accuracy of 92.8%.
Citing this Journal Article :Linfeng Sui, Xuyang Zhao, Jianting Cao, Qibin Zhao, "Localization of epileptic foci from ieeg via mixed convolutional neural network", Volume 14 Issue 4 - October 2019, 8-13
Click here to Submit Copyright Takedown Notice for this article.