EEG signal multi-classification method and system based on EEGNet and LSTM parallel network

• Main Authors: LUO XIANG, HU XUANYE
• Format: Patent
• Language: Chinese; English
• Published: 22.09.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DIAGNOSIS; ELECTRIC DIGITAL DATA PROCESSING; HUMAN NECESSITIES; HYGIENE; IDENTIFICATION; MEDICAL OR VETERINARY SCIENCE; PHYSICS; SURGERY

The invention provides an EEG signal multi-classification method and system based on an EEGNet and LSTM parallel network, and the method comprises the following steps: constructing an EEG-LSTMNet model which comprises an EEGNet module and an LSTM module which are arranged in parallel; parameter setting and training are carried out on the EEG-LSTMNet model, and a trained EEG-LSTMNet model is obtained; electroencephalogram signals are obtained, feature extraction is conducted on the electroencephalogram signals through an EEGNet module and an LSTM module in the trained EEG-LSTMNet model, and EEGNet frequency domain features and LSTM time domain features are obtained; the EEGNet frequency domain features and the LSTM time domain features are fused, and a final electroencephalogram signal classification result is obtained. According to the method, the EEG-LSTMNet model is constructed, the EEGNet and the LSTM are arranged in parallel, the EEGNet inherits the advantages of the CNN, the defect that the same CNN cann