Prognosis of Sleep Stage Classification Using Machine Learning Techniques Applied on Single-channel of EEG signal of both Healthy Subjects and Mild Sleep effected Subjects

Sleep is a basic requirement of human life. It is one of the vital roles in to the human life to maintain the proper mental health, physical health and quality of life. In this proposed research work, we conduct an automated sleep stage classification to proper investigation of irregularities occurred during sleep based on single channel of electroencephalogram (EEG) signal (SleepEEG) with using of machine learning approaches. The major advantage of this proposed research work over standard polysomnography method are: 1) it measures the sleep irregularities during sleep by considering two different medical condition subjects of different gender with different age groups.2) One more important objective of this proposed sleep study is that here we obtain different session recordings to investigate on sleep abnormality patterns, which can help to find better diagnosis towards treatment of sleep related disorder.3)In present work, we have obtained 15s time-framework epochs from individual subjects to check which window size is more effective towards identification on sleep irregularities.The present research work based on two-state sleep stage classification problem based on single channel of EEG signal were performed in different stepwise manner such as acquisition of data from participated subjects, preprocessing, feature extraction,feature selection and classification. We obtained the EEG data from ISRUC-Sleep data repository for measuring the performances of the proposed framework, where the sleep stages are visually labelled. The obtained results demonstrated that the proposed methodologies achieves high classification accuracy, which support to sleep experts for accurately measure the irregularities occurred during sleep and also helps the clinicians to evaluate the presence and criticality of sleep related disorders.