Automatically Detected Microsleep Episodes in the Fitness-to-Drive Assessment

Association of Visual-Based Signals with Electroencephalography Patterns in Enhancing the Drowsiness Detection in Drivers with Obstructive Sleep Apnea

Individuals with obstructive sleep apnea (OSA) face increased accident risks due to excessive daytime sleepiness. PERCLOS, a recognized drowsiness detection method, encounters challenges from image quality, eyewear interference, and lighting variations, impacting its performance, and requiring validation through physiological signals. We propose visual-based scoring using adaptive thresholding for eye aspect ratio with OpenCV for face detection and Dlib for eye detection from video recordings. This technique identified 453 drowsiness (PERCLOS ≥ 0.3 || CLOSDUR ≥ 2 s) and 474 wakefulness episodes (PERCLOS < 0.3 and CLOSDUR < 2 s) among fifty OSA drivers in a 50 min driving simulation while wearing six-channel EEG electrodes. Applying discrete wavelet transform, we derived ten EEG features, correlated them with visual-based episodes using various criteria, and assessed the sensitivity of brain regions and individual EEG channels. Among these features, theta-alpha-ratio exhibited robust mapping (94.7%) with visual-based scoring, followed by delta-alpha-ratio (87.2%) and delta-theta-ratio (86.7%). Frontal area (86.4%) and channel F4 (75.4%) aligned most episodes with theta-alpha-ratio, while frontal, and occipital regions, particularly channels F4 and O2, displayed superior alignment across multiple features. Adding frontal or occipital channels could correlate all episodes with EEG patterns, reducing hardware needs. Our work could potentially enhance real-time drowsiness detection reliability and assess fitness to drive in OSA drivers.

Embracing sleep-onset complexity

Sleep is crucial for many vital functions and has been extensively studied. By contrast, the sleep-onset period (SOP), often portrayed as a mere prelude to sleep, has been largely overlooked and remains poorly characterized. Recent findings, however, have reignited interest in this transitional period and have shed light on its neural mechanisms, cognitive dynamics, and clinical implications. This review synthesizes the existing knowledge about the SOP in humans. We first examine the current definition of the SOP and its limits, and consider the dynamic and complex electrophysiological changes that accompany the descent to sleep. We then describe the interplay between internal and external processing during the wake-to-sleep transition. Finally, we discuss the putative cognitive benefits of the SOP and identify novel directions to better diagnose sleep-onset disorders.

Sleepiness and the transition from wakefulness to sleep

The transition from wakefulness to sleep is a progressive process that is reflected in the gradual loss of responsiveness, an alteration of cognitive functions, and a drastic shift in brain dynamics. These changes do not occur all at once. The sleep onset period (SOP) refers here to this period of transition between wakefulness and sleep. For example, although transitions of brain activity at sleep onset can occur within seconds in a given brain region, these changes occur at different time points across the brain, resulting in a SOP that can last several minutes. Likewise, the transition to sleep impacts cognitive and behavioral levels in a graded and staged fashion. It is often accompanied and preceded by a sensation of drowsiness and the subjective feeling of a need for sleep, also associated with specific physiological and behavioral signatures. To better characterize fluctuations in vigilance and the SOP, a multidimensional approach is thus warranted. Such a multidimensional approach could mitigate important limitations in the current classification of sleep, leading ultimately to better diagnoses and treatments of individuals with sleep and/or vigilance disorders. These insights could also be translated in real-life settings to either facilitate sleep onset in individuals with sleep difficulties or, on the contrary, prevent or control inappropriate sleep onsets.

Dashcam video footage-based analysis of microsleep-related behaviors in truck collisions attributed to falling asleep at the wheel

OBJECTIVE

With the rapid spread of dashcams, many car accidents have been recorded; however, behavioral approaches using these dashcam video footage have not been sufficiently examined. We employed dashcam video footage to evaluate microsleep-related behaviors immediately prior to real-world truck collisions in professional drivers to explore a new solution to reduce collisions attributed to falling asleep at the wheel.

METHODS

In total, 3,120 s of video footage (60 s/case × 52 cases) from real-world truck collisions of 52 professional drivers obtained from interior and exterior dashcams were used and visually analyzed in a second-by-second manner to simultaneously evaluate any eye changes and microsleep-related behaviors (the driver's anti-sleepiness behavior, behavioral signs of microsleep, and abnormal vehicle behavior) during driving.

RESULTS

Assessment of the frequency of occurrence of each item of microsleep-related behavior in the 52 collisions revealed that the item "touching" in terms of anti-sleepiness behavior, "absence of body movement" in terms of behavioral signs of microsleep, and "inappropriate line crossing" in terms of abnormal vehicle behavior were observed at the highest rate in all drivers (46.2%, 75.0%, and 78.8%, respectively). Decreases in anti-sleepiness behavior coincided with increases in behavioral signs of microsleep and abnormal vehicle behavior, with collisions occurring within approximately 40 s of these changes. Collisions were more common among young people and in the early morning and evening.

CONCLUSION

Our dashcam video footage-based analysis in truck collisions attributed to falling asleep at the wheel revealed the process of changes in microsleep-related driver and vehicle behaviors, classified as anti-sleepiness behavior, behavioral signs of microsleep, and abnormal vehicle behavior. Based on these findings, to prevent collisions caused by falling asleep at the wheel, it is crucial to monitor not only the driver's eyes, but also the driver's whole body and vehicle behavior simultaneously to reliably detect microsleep-related behaviors.

2020 International brain-computer interface competition: A review

The brain-computer interface (BCI) has been investigated as a form of communication tool between the brain and external devices. BCIs have been extended beyond communication and control over the years. The 2020 international BCI competition aimed to provide high-quality neuroscientific data for open access that could be used to evaluate the current degree of technical advances in BCI. Although there are a variety of remaining challenges for future BCI advances, we discuss some of more recent application directions: (i) few-shot EEG learning, (ii) micro-sleep detection (iii) imagined speech decoding, (iv) cross-session classification, and (v) EEG(+ear-EEG) detection in an ambulatory environment. Not only did scientists from the BCI field compete, but scholars with a broad variety of backgrounds and nationalities participated in the competition to address these challenges. Each dataset was prepared and separated into three data that were released to the competitors in the form of training and validation sets followed by a test set. Remarkable BCI advances were identified through the 2020 competition and indicated some trends of interest to BCI researchers.

Epidemiology, Physiology and Clinical Approach to Sleepiness at the Wheel in OSA Patients: A Narrative Review

Sleepiness at the wheel (SW) is recognized as an important factor contributing to road traffic accidents, since up to 30 percent of fatal accidents have been attributed to SW. Sleepiness-related motor vehicle accidents may occur both from falling asleep while driving and from behavior impairment attributable to sleepiness. SW can be caused by various sleep disorders but also by behavioral factors such as sleep deprivation, shift work and non-restorative sleep, as well as chronic disease or the treatment with drugs that negatively affect the level of vigilance. An association between obstructive sleep apnea (OSA) and motor vehicle accidents has been found, with an increasing risk in OSA patients up to sevenfold in comparison to the general population. Regular treatment with continuous positive airway pressure (CPAP) relieves excessive daytime sleepiness and reduces the crash risk. Open questions still remain about the physiological and clinical determinants of SW in OSA patients: the severity of OSA in terms of the frequency of respiratory events (apnea hypopnea index, AHI) or hypoxic load, the severity of daytime sleepiness, concomitant chronic sleep deprivation, comorbidities, the presence of depressive symptoms or chronic fatigue. Herein, we provide a review addressing the epidemiological, physiological and clinical aspects of SW, with a particular focus on the methods to recognize those patients at risk of SW.

Predicting the loss of responsiveness when falling asleep in humans

Falling asleep is a dynamical process that is poorly defined. The period preceding sleep, characterized by the progressive alteration of behavioral responses to the environment, which may last several minutes, has no electrophysiological definition, and is embedded in the first stage of sleep (N1). We aimed at better characterizing this drowsiness period looking for neurophysiological predictors of responsiveness using electro and magnetoencephalography. Healthy participants were recorded when falling asleep, while they were presented with continuous auditory stimulations and asked to respond to deviant sounds. We analysed brain responses to sounds and markers of ongoing activity, such as information and connectivity measures, in relation to rapid fluctuations of brain rhythms observed at brain onset and participants' capabilities to respond. Results reveal a drowsiness period distinct from wakefulness and sleep, from alpha rhythms to the first sleep spindles, characterized by diverse and transient brain states that come on and off at the scale of a few seconds and closely reflects, mainly through neural processes in alpha and theta bands, decreasing probabilities to be responsive to external stimuli. Results also show that the global P300 was only present in responsive trials, regardless of vigilance states. A better consideration of the drowsiness period through a formalized classification and its specific brain markers such as described here should lead to significant advances in vigilance assessment in the future, in medicine and ecological environments.

Microsleep assessment enhances interpretation of the Maintenance of Wakefulness Test

STUDY OBJECTIVES

The Maintenance of Wakefulness Test (MWT) is used to objectively evaluate an individual's ability to remain awake; however, microsleeps are not included in the assessment. We aimed to determine if microsleep data prior to sleep onset assisted in interpretation of ability to maintain wakefulness across a range of typical patient groups.

METHODS

Forty-eight patients referred for overnight polysomnography and subsequent MWT were included. Patients were divided into 3 groups (treated obstructive sleep apnea [OSA], untreated OSA, or treated idiopathic hypersomnia or narcolepsy) based on prior medical diagnosis. Demographics, clinical characteristics, polysomnography, and MWT variables, including frequency, distribution, duration, and latency of microsleeps were compared between groups.

RESULTS

Microsleeps were observed in MWT trials significantly more frequently in patients with treated idiopathic hypersomnia/narcolepsy over the course of the day (0.34 ± 0.06 vs 0.07 ± 0.02 microsleeps/min; P < .001) and in patients with untreated OSA toward the end of the day (0.31 ± 0.06 vs 0.05 ± 0.02 microsleeps/min; P < .001) compared to the group with treated OSA. Microsleeps were often observed in series and earlier in patients with treated idiopathic hypersomnia/narcolepsy (10.9 ± 1.6 minutes) and those with untreated OSA (16.2 ± 2.7 minutes) compared to the group with treated OSA (24.9 ± 3.0 minutes; P < .05), and, if taken into consideration, would increase the proportion of patients demonstrating inability to maintain wakefulness by 33% and 22%, respectively.

CONCLUSIONS

MWT performance varies significantly across patient groups. Microsleep analysis prior to sleep onset may be a more sensitive measure of patient daytime wakefulness than sleep latency alone and should be considered in MWT assessment.

CITATION

Anniss AM, Young A, O'Driscoll DM. Microsleep assessment enhances interpretation of the Maintenance of Wakefulness Test. J Clin Sleep Med. 2021;17(8):1571-1578.

Towards better microsleep predictions in fatigued drivers: exploring benefits of personality traits and IQ

Fatigued driving is one of the main contributors to road traffic accidents. Poor sleep quality and lack of sleep negatively affect driving performance, and extreme states of fatigue can cause microsleep (i.e., short episodes of sleep with complete loss of awareness). Driver monitoring systems analyse biosignals (e.g., gaze, blinking, heart rate) and vehicle data (e.g., steering wheel movements, lane holding, acceleration) to detect states of fatigue and prevent accidents. We argue that inter-individual differences in personality, sensation seeking behaviour, and intelligence could improve microsleep prediction, in addition to sleepiness. We tested 144 male participants in a supervised driving track after 27 hours of sleep deprivation. More than 74% of drivers experienced microsleep, after an average driving time of 52 min. Overall, prediction models for microsleep vulnerability and driving time before microsleep were significantly improved by conscientiousness, sensation seeking and non-verbal IQ, in addition to situational sleepiness, as individual risk factors. Practitioner summary: This study offers valuable insights for the design of driver monitoring systems. The use of individual risk factors such as conscientiousness, sensation seeking, and non-verbal IQ can increase microsleep prediction. These findings may improve monitoring systems based solely on physiological signals (e.g., blinking, heart rate) and vehicle data (e.g., steering wheel movements, acceleration, cornering). Abbreviations: ADAC: Allgemeiner Deutscher Automobil Club; ANOVA: analysis of variance; AIC: Akaike information criteria; CI: confidence interval; GPS: global positioning system; IQ: intelligence quotient; IQR: inter quartile range; KSS: Karolinska sleepiness scale; NEO-PI-R: revised NEO personality inventory; OLS: ordinary least squares; PSQI: Pittsburgh sleep quality index; SPM: standard progressive matrices; SSS: sensation seeking scale; WHO: World Health Organization.

Microsleep versus Sleep Onset Latency during Maintenance Wakefulness Tests: Which One Is the Best Marker of Sleepiness?

The interpretation of the Maintenance Wakefulness Test (MWT) relies on sleep onset detection. However, microsleeps (MSs), i.e., brief periods of sleep intrusion during wakefulness, may occur before sleep onset. We assessed the prevalence of MSs during the MWT and their contribution to the diagnosis of residual sleepiness in patients treated for obstructive sleep apnea (OSA) or hypersomnia. The MWT of 98 patients (89 OSA, 82.6% male) were analyzed for MS scoring. Polysomnography parameters and clinical data were collected. The diagnostic value for detecting sleepiness (Epworth Sleepiness Scale > 10) of sleep onset latency (SOL) and of the first MS latency (MSL) was assessed by the area under the receiver operating characteristic (ROC) curve (AUC, 95% CI). At least one MS was observed in 62.2% of patients. MSL was positively correlated with SOL (r = 0.72, p < 0.0001) but not with subjective scales, clinical variables, or polysomnography parameters. The use of SOL or MSL did not influence the diagnostic performance of the MWT for subjective sleepiness assessment (AUC = 0.66 95% CI (0.56, 0.77) versus 0.63 95% CI (0.51, 0.74)). MSs are frequent during MWTs performed in patients treated for sleep disorders, even in the absence of subjective sleepiness, and may represent physiological markers of the wake-to-sleep transition.

Automatic Detection of Microsleep Episodes With Deep Learning

Brief fragments of sleep shorter than 15 s are defined as microsleep episodes (MSEs), often subjectively perceived as sleepiness. Their main characteristic is a slowing in frequency in the electroencephalogram (EEG), similar to stage N1 sleep according to standard criteria. The maintenance of wakefulness test (MWT) is often used in a clinical setting to assess vigilance. Scoring of the MWT in most sleep-wake centers is limited to classical definition of sleep (30 s epochs), and MSEs are mostly not considered in the absence of established scoring criteria defining MSEs but also because of the laborious work. We aimed for automatic detection of MSEs with machine learning, i.e., with deep learning based on raw EEG and EOG data as input. We analyzed MWT data of 76 patients. Experts visually scored wakefulness, and according to recently developed scoring criteria MSEs, microsleep episode candidates (MSEc), and episodes of drowsiness (ED). We implemented segmentation algorithms based on convolutional neural networks (CNNs) and a combination of a CNN with a long-short term memory (LSTM) network. A LSTM network is a type of a recurrent neural network which has a memory for past events and takes them into account. Data of 53 patients were used for training of the classifiers, 12 for validation and 11 for testing. Our algorithms showed a good performance close to human experts. The detection was very good for wakefulness and MSEs and poor for MSEc and ED, similar to the low inter-expert reliability for these borderline segments. We performed a visualization of the internal representation of the data by the artificial neuronal network performing best using t-distributed stochastic neighbor embedding (t-SNE). Visualization revealed that MSEs and wakefulness were mostly separable, though not entirely, and MSEc and ED largely intersected with the two main classes. We provide a proof of principle that it is feasible to reliably detect MSEs with deep neuronal networks based on raw EEG and EOG data with a performance close to that of human experts. The code of the algorithms (https://github.com/alexander-malafeev/microsleep-detection) and data (https://zenodo.org/record/3251716) are available.