An eeg emotion classification system based on one-dimension convolutional neural networks and virtual reality

Abstract

The increase in the number of patients with Alzheimer’s disease has placed a heavy burden on society and has become a major problem in the medical field. In recent years, brain-computer interface (BCI) has become an important way to explore the utilization of modern technology to improve Alzheimer’s disease. However, traditional pattern recognition methods suffer from poor classification and feature extraction. This paper proposes an end-to-end Electroencephalograph (EEG) emotion classification method based on 1D-CNN (one-dimension convolutional neural networks) to improve the accuracy of BCI pattern recognition. An application system combining virtual reality (VR) and BCI is further constructed, which could help patients to repeatedly perform memory stimulation and provide a new clue for clinical treatment of Alzheimer’s disease.