A comprehensive evaluation of contemporary methods used for automatic sleep staging. (August 2022)

• Record Type: Journal Article
• Authors: Sarkar, Debanjan; Guha, Dwaipayan; Tarafdar, Puspal; Sarkar, Shusovan; Ghosh, Antara; Dey, Debangshu
• Is Part Of: Biomedical signal processing and control. Volume 77(2022)
• Publication Date: 2022-08
• Subjects: Polysomnography -- Electroencephalography -- Rapid eye movement -- Bilateral long short-term memory -- Machine learning -- Deep learning; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2022.103819
• Languages: English
• ISSNs: 1746-8094

Abstract

Abstract: This paper presents a systematic review of automatic sleep staging studies. Polysomnographic (PSG) data is used for the study of sleep staging. The benchmark for sleep staging involves the division of full-night PSG into 30-s epochs and then scoring them manually, according to the guidelines set by the American Academy of Sleep Medicine (AASM). Accurate scoring of sleep can help diagnose physiological and neurological disorders in patients. Automatic sleep stage classification (ASSC) has been a major topic in the field of biomedical signal processing for over two decades. It has the potential to provide a cost effective, faster and more accessible medium for evaluating sleep patterns and monitoring sleep quality compared to manual scoring. The procedure for automatic sleep staging can be broadly divided into two steps: feature extraction and sleep stage classification. With the developments in the fields of signal processing, a wide variety of signal pre-processing methods have been adopted by researchers leading to better feature extraction and selection. This review article mainly focuses on sleep stage classification using machine learning models and different types of neural networks.