IEEG-CT: A CNN and Transformer Based Method for Intracranial EEG Signal Classification

Intracranial electroencephalography (iEEG) is of great importance for the preoperative evaluation of drug-resistant epilepsy. Automatic classification of iEEG signals can speed up the process of epilepsy diagnosis. Existing deep learning-based approaches for iEEG signal classification usually rely on convolutional neural network (CNN) and long short-term memory network. However, these approaches have limitations in terms of classification accuracy. In this study, we propose a CNN and Transformer based method, which is named as IEEG-CT, for iEEG signal classification. Firstly, IEEG-CT utilizes deep one-dimensional CNN to extract the critical local features from the raw iEEG signals. Secondly, IEEG-CT combines a Transformer encoder, which employs a multi-head attention mechanism to capture long-range global information among the extracted features. In particular, we leverage a causal convolution multi-head attention instead of the standard Transformer block to efficiently capture the temporal dependencies within the input features. Finally, the obtained global features by the Transformer encoder are employed for the classification. We assess the performance of IEEG-CT on two publicly available multicenter iEEG datasets. According to the experimental results, IEEG-CT surpasses state-of-the-art techniques in terms of several evaluation metrics, i.e., accuracy, AUROC, and AUPRC.