A scheme combining feature fusion and hybrid deep learning models for epileptic seizure detection and prediction

• Published in: Scientific reports Vol. 14; no. 1; pp. 16916 - 16
• Main Authors: Zhang, Jincan, Zheng, Shaojie, Chen, Wenna, Du, Ganqin, Fu, Qizhi, Jiang, Hongwei
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 23.07.2024; Nature Publishing Group; Nature Portfolio
• Subjects: 692/308; 692/617; Algorithms; Attention mechanism (AM); Convolutional neural network (CNN); Convulsions & seizures; Deep Learning; EEG; Electroencephalography (EEG) signal; Electroencephalography - methods; Epilepsy; Epilepsy - diagnosis; Epilepsy - physiopathology; Gated recurrent unit (GRU); Humanities and Social Sciences; Humans; Multi-class feature fusion; multidisciplinary; Neural networks; Neural Networks, Computer; Neurological diseases; Predictions; Quality of life; Science; Science (multidisciplinary); Seizure detection and prediction; Seizures; Seizures - diagnosis; Signal Processing, Computer-Assisted; Wavelet Analysis; Wavelet transforms; Attention mechanism (AM); Gated recurrent unit (GRU); Convolutional neural network (CNN); Seizure detection and prediction; Electroencephalography (EEG) signal; Multi-class feature fusion
• DOI: 10.1038/s41598-024-67855-4
• ISSN: 2045-2322

Epilepsy is one of the most well-known neurological disorders globally, leading to individuals experiencing sudden seizures and significantly impacting their quality of life. Hence, there is an urgent necessity for an efficient method to detect and predict seizures in order to mitigate the risks faced by epilepsy patients. In this paper, a new method for seizure detection and prediction is proposed, which is based on multi-class feature fusion and the convolutional neural network-gated recurrent unit-attention mechanism (CNN-GRU-AM) model. Initially, the Electroencephalography (EEG) signal undergoes wavelet decomposition through the Discrete Wavelet Transform (DWT), resulting in six subbands. Subsequently, time–frequency domain and nonlinear features are extracted from each subband. Finally, the CNN-GRU-AM further extracts features and performs classification. The CHB-MIT dataset is used to validate the proposed approach. The results of tenfold cross validation show that our method achieved a sensitivity of 99.24% and 95.47%, specificity of 99.51% and 94.93%, accuracy of 99.35% and 95.16%, and an AUC of 99.34% and 95.15% in seizure detection and prediction tasks, respectively. The results show that the method proposed in this paper can effectively achieve high-precision detection and prediction of seizures, so as to remind patients and doctors to take timely protective measures.