A Hybrid Deep Learning Scheme for Multi-Channel Sleep Stage Classification

Sleep stage classification plays a significant role in the accurate diagnosis and treatment of sleep-related diseases. This study aims to develop an efficient deep learning based scheme for correctly identifying sleep stages using multi-biological signals such as electroencephalography (EEG), electrocardiogram (ECG), electromyogram (EMG), electrooculogram (EOG). Most prior studies focus on hand-crafted feature extraction methods. Traditional methods choose features manually from raw data, which is tedious, these are limited their ability balance efficiency accuracy. Moreover, most existing works staging either single channel (a single-lead EEG may not contain enough information) or only signal can reveal more complicated physical reliable various stages. proposes approach combine Convolutional Neural Networks (CNNs) Gated Recurrent Units (GRUs) that discover hidden data recognize different efficiently. In proposed scheme, CNN designed extract concealed signals, GRU employed automatically learn transition rules among After that, softmax layers used classify The method was tested two publicly available databases: Heart Health Study (SHHS) St. Vincent's University Hospital/University College Dublin Apnoea (UCDDB). experimental results model yields better performance compared state-of-the-art works. Our will assist building new system deal with multi-channel multi-modal processing tasks applications.