A Convolutional Neural Network Based on Raw Single Channel EEG for Automatic Sleep Staging

Abstract

Sleep stages are determined firstly for the evaluation of sleep quality and the diagnosis of sleep diseases. The signals, recorded from sensors connected to various parts of the body, such as electroencephalogram (EEG), electrocardiogram (ECG), electrooculogram (EOG) and electromyogram (EMG) are used for this purpose. After the production of affordable wearable EEG devices for individual use, studies have begun to detect sleep stages from a single channel EEG signal. This paper presents an automated system that can perform sleep staging using a single-channel raw EEG signal. A Convolutional Neural Network (CNN) model was trained with the raw EEG signal for sleep stage detection. The use of CNN does not require any feature extraction. The developed CNN model classifies the sleep data sampled at 250 Hz, divided into 30-second segments according to the 5-class sleep staging system. According to the test results, the performance of the proposed system was found to be 93% macro F1 score and 92% accuracy.