Current status and prospects of automatic sleep stages scoring: Review

Abstract

The scoring of sleep stages is one of the essential tasks in sleep analysis. Since a manual procedure requires considerable human and financial resources, and incorporates some subjectivity, an automated approach could result in several advantages. There have been many developments in this area, and in order to provide a comprehensive overview, it is essential to review relevant recent works and summarise the characteristics of the approaches, which is the main aim of this article. To achieve it, we examined articles published between 2018 and 2022 that dealt with the automated scoring of sleep stages. In the final selection for in-depth analysis, 125 articles were included after reviewing a total of 515 publications. The results revealed that automatic scoring demonstrates good quality (with Cohen's kappa up to over 0.80 and accuracy up to over 90%) in analysing EEG/EEG + EOG + EMG signals. At the same time, it should be noted that there has been no breakthrough in the quality of results using these signals in recent years. Systems involving other signals that could potentially be acquired more conveniently for the user (e.g. respiratory, cardiac or movement signals) remain more challenging in the implementation with a high level of reliability but have considerable innovation capability. In general, automatic sleep stage scoring has excellent potential to assist medical professionals while providing an objective assessment.

NAPping PAnts (NAPPA): An open wearable solution for monitoring Infant's sleeping rhythms, respiration and posture

“…In comparison, the reported classification accuracy from the NAPPA-based classifier is surprisingly high when considering it comes without measures of cortical, muscular, or ocular activity. This is best reconciled in the context of more recent work, including the present findings, which show that many alternative sets of physiological parameters may provide reasonably accurate sleep classifications [ 21 , 56 ], and the choice of signals needs to be tailored concerning specific, practical considerations in each use case, such as the casual home recordings in young infants. While some of the apparent epoch-level misclassifications are genuine classification errors, it is also essential to recognize that sleep states are not biologically discrete [ 51 , 52 , 57 ]; our classifier output, SDT, recognizes this graded nature of sleep; its actual correspondence to the real infant's sleep calls for benchmarks other than conventional, discrete sleep scores.…”

“…In the recent literature, several new non-invasive solutions have been proposed for other types of out-of-hospital recordings of infant sleep, including physiological recordings of respiration variability, electroencephalography, body movements, heart rate, and video [ [16] , [17] , [18] , [19] ]. Analysis of such recordings can be automated to support the development of sleep state classifier algorithms [ 14 , 20 , 21 ], such as those based on electroencephalography (EEG) [ 22 , 23 ], electrocardiography (ECG) [ 24 , 25 ], and respiration signals [ 18 , 24 , [26] , [27] , [28] ], or video-based assessments of the same physiological functions by using auto-videosomnography approach [ 19 , 29 ]. Even movement signals as simple as a bed mattress or accelerometer can provide reasonably accurate sleep state classification when using tailored computational measures of respiration, posture, and body movements [ 18 , 30 , 31 ].…”

Automatic sleep scoring for real-time monitoring and stimulation in individuals with and without sleep apnea

“…In clinics, sleep doctors could integrate automatic sleep scoring tools to streamline the diagnostic process and facilitate timely interventions. By automating the labor-intensive task of sleep staging, clinicians could allocate more time to patient care and decision-making, ultimately improving the quality and efficiency of healthcare delivery (Gaiduk et al 2023; Soleimani et al 2023). Moreover, the utility of automatic sleep scoring extends beyond the clinic and home-based monitoring may facilitate earlier detection of sleep disorders.…”

SleepEEGpy: a Python-based software “wrapper” package to organize preprocessing, analysis, and visualization of sleep EEG data

“…EEG is routinely used in conjunction with electrooculography and electromyography to perform sleep scoring and distinguish between vigilance states of wakefulness, rapid eye movement (REM) sleep, and non-REM sleep (2). Sleep scoring is performed either manually according to established standards (3) or, in recent years, via automatic tools (4)(5)(6). Beyond sleep scoring, there is increased attention toward advanced EEG analysis that focuses on investigating events occurring at specific times, frequencies, and scalp locations or in estimated brain sources (7).…”