Article Preview
TopIntroduction
Maintaining proper health and mental stability is critical for overall health and well-being. Despite a good deal of research investment, sleep quality continues to be a crucial public challenge. Nowadays, people of all age groups are affected by improper sleep quality. Poor sleep can lead to a variety of neurological disorders (Panossian, 2009; Smaldone, 2007). Sleep disorders are common in all subsets of the population, independently of gender. This public health challenge greatly affects quality of life in terms of both physical and mental health. Insomnia, parasomnias, sleep-related breathing difficulties, hypersomnia, bruxism, narcolepsy, and circadian rhythm disorders are some common examples of sleep-related disorders. Some of these disorders can be treated with proper analysis of early symptoms; in such cases, adequate sleep quality is essential for the patient’s recovery. Moreover, numerous sleep disorders can be clinically diagnosed with the help of computer-aided technologies (Hassan, 2016). Sleep monitoring is one of the most significant activities in the assessment of sleep-related disturbances and other neural problems. Sleep is a dynamic process and includes different sleep stages, including the waking, non-rapid eye movement (NREM), and rapid eye movement (REM) stages. Furthermore, NREM sleep state is divided into four stages, namely NREM N1, N2, N3, and N4 (Aboalayon, 2014). The wake stage is the period of awakening before sleep. The NREM sleep stages are sequentially indicative of light to deep sleep. N1 is a light sleep stage with slow eye and muscle movements. True sleep begins with stage N2, where eye movements stop and brain activity decreases. The N3 and N4 stages are periods of deep sleep without eye and muscle movements. Finally, in the REM stage, rapid eye movements occur and breathing increases. A nightly sleep cycle consists of approximately 75% NREM sleep and 25% REM sleep (Obayya, 2014).A sleep assessment can be supported by a sleep test with polysomnographic (PSG) recordings. PSG signals are a collection of different physiological signals that are collected from subjects during sleep.A PSG signal is combination of multivariate signal recordings, such as electroencephalogram (EEG), electrocardiogram (ECG), electrooculogram (EOG), and electromyogram (EMG) (Alickovic,2018) . The EEG signal recordings are used during sleep staging scoring. These signals represent brain activity, and, therefore, are suitable for evaluation of sleep abnormalities. After data collection, a sleep staging score is given. The recorded EEG signals are extracted through multiple fixed electrodes located in different places on a patient’s scalp. The process of electrode placement is done according to the international 10/20 placement system (Abeyratne, 2007).The entire process is carried out by sleep experts who analyze the different patterns of sleep states. The evaluation is made through visual inspection using the recorded data for a specific time window. Consequently, the sleep score is determined through multiple criteria. The criteria for the sleep scoring process are based on the Rechtschaffen Kales guidelines (Rechtschaffen, 1968). These guidelines classify the sleep stages as wake (W), non-rapid eye movement (N1, N2, N3, N4), and rapid eye movement (REM). The proposed guidelines also include minor changes introduced by the American Academy of Sleep Medicine (AASM) (Iber, 2007). The AASM manuals have combined the N3 and N4 stages into a single stage (N3) that is characterized by slow-wave sleep (SWS). Manual and visual evaluation of sleep scoring is complicated, costly and time-consuming. This manual approach overloads sleep experts who have to continuously monitor their patients during every sleep evaluation. However, this process provides the best accuracy in research on sleep disturbances (Bianchi, 2017). Consequently, the development of automated detection and recognition applications to assist sleep experts with the process of diagnosing sleep disorders is critical for enhanced public health. The EEG signals that are collected during sleep studies consist of either single-channel or multiple-channel recordings. EEG signals have several advantages over other types of signals. EEG signals can be obtained using wearable technologies that is comfortable for subjects. Moreover, the data collection process can be done in either the patient’s home or a healthcare facility (Cogan, 2017).