Dimension reduction based on Canonical Correlation Analysis technique to classify sleep stages of sleep apnea disorder using EEG and ECG signals

Sleep stage scoring is the first step to diagnostic of sleep disorders and it is scored by the conventional method known as the visual sleep stage scoring based on human. To assist the sleep physician in evaluating of patients, a new automatic sleep stage classification system needs to be developed. So this is the aim of this work based on Electroencephalography (EEG) and Electrocardiography (ECG) for Sleep apnea patients. This article proposes two importance topics, the first is the new feature of EEG signal using a simple statistical technique and the results prove that the various sleep stages can be discriminated more clearly at significant levels (p<<0.05). Second, the dimension reduction is proposed based on the Canonical Correlation Analysis (CCA) technique that explores possible correlated multi-sources to improve the sleep stages classification at 95.42% accuracy by using random forest classification. The results show that our proposed method has ability to develop a new sleep stage classification assistance.