Sleep posture recognition based on machine learning: A systematic review

Background:In recent years, the application of artificial intelligence in the field of sleep medicine has rapidly emerged. One of the main concerns of many researchers is the recognition of sleep positions, which enables efficient monitoring of changes in sleeping posture for precise and intelligent adjustment. In sleep monitoring, machine learning is able to analyze the raw data collected and optimizes the algorithm in real-time to recognize the sleeping position of the human body during sleep.Methodology:A detailed search of relevant databases was conducted through a systematic search process, and we reviewed research published since 2017, focusing on 27 articles on sleep recognition.Results:Through the analysis and study of these articles, we propose several determinants that objectively affect sleeping posture recognition, including the acquisition of sleep posture data, data pre-processing, recognition algorithms, and validation analysis. Moreover, we analyze the categories of sleeping postures adapted to different body types.Conclusion:A systematic evaluation combining the above determinants provides solutions for system design and rational selection of recognition algorithms for sleep posture recognition, and it is necessary to regularize and standardize existing machine learning algorithms before they can be incorporated into clinical monitoring of sleep.     

Neural Network Model: Application to Automatic Analysis of Human Sleep

We describe an approach to automatic all-night sleep analysis based on neural network models and simulated on a digital computer. First, automatic sleep stage scoring was performed using a multilayer feedforward network. Second, supervision of the automatic decision was achieved using ambiguity rejection and artifact rejection. Then, numerical analysis of sleep was carried out using all-night spectral analysis for the background activity of the EEG and sleep pattern detectors for the transient activity. Computerized analysis of sleep recordings may be considered as an essential tool to describe the sleep process and to reflect the dynamical organization of human sleep.     

Automated detection of cyclic alternating pattern and classification of sleep stages using deep neural network

The visual sleep stages scoring by human experts is the current gold standard for sleep analysis. However, this method is tedious, time-consuming, prone to human errors, and unable to detect microstructure of sleep such as cyclic alternating pattern (CAP) which is an important diagnostic factor for the detection of sleep disorders such as insomnia and obstructive sleep apnea (OSA). The CAP is only observed as subtle changes in the electroencephalogram (EEG) signals during non-rapid eye movement (NREM) sleep, making it very difficult for human experts to discern. Hence, it is important to have an automated system developed using artificial intelligence for accurate and robust detection of CAP and sleep stages classification. In this study, a deep learning model based on 1-dimensional convolutional neural network (1D-CNN) is proposed for CAP detection and homogenous 3-class sleep stages classification, namely wakefulness (W), rapid eye movement (REM) and NREM sleep. The proposed model is developed using standardized EEG recordings. Our developed CNN network achieved good model performance for 3-class sleep stages classification with a classification accuracy of 90.46%. Our proposed model also yielded a classification accuracy of 73.64% using balanced CAP dataset, and sensitivity of 92.06% with unbalanced CAP dataset. Our proposed model correctly identified majority of A-phases which comprised of only 12.6% in the unbalanced dataset. The performance of the developed prototype is ready to be tested with more data before clinical application.

     

Autonomic function assessment using analysis of heart rate variability

Analysis of changes in heart rate can be useful in determining the state of various body systems. In particular the analysis of heart rate variability (HRV) is used in the assessment of autonomic function. This paper uses the discrete harmonic wavelet transform for a time-frequency analysis of HRV data to show changes in spectral power over time. Signals representing patient heart rate are presented, and methods for spectral and time-frequency analysis are described. Three sets of patient data are then analysed using these methods. The results show the potential of time-frequency analysis in the assessment of medical disorders, such as the sleep apnoea syndrome, where transient alterations in autonomic function occur.     

Computer program for automated sleep depth estimation

In this article, we present a new implementation of an amplitude-independent method for continuous-scale sleep depth estimation. Having been implemented as an add-on analysis module under commercially available biosignal recording and analysis software, it can be easily applied in clinical routine. The software gives the user full freedom to change all the analysis parameters inside theoretical limits. Computational sleep depth profiles produced by the presented software compare favourably with visual classifications. Future work will concentrate on systematic optimization of analysis parameters, further evaluation of the method with disturbed sleep and application of the method for automated adaptive sleep analysis.     

A polysomnographic study of morning and afternoon sleep quality during simulated night shift work

Shiftworkers differ as to scheduling their day sleep between night shifts. An experimental study was carried out to compare the effects of morning and afternoon sleep behaviour on sleep quality within and after a period of seven night shifts. Twelve young males participated in a repeated–measures design, six sleeping in the morning and six in the afternoon after night shifts. Sleep quality was evaluated by polysomnographic and subjective measures. Morning and afternoon sleepers showed some differences in the trend of sleep changes over successive sleep periods. However, at the end of the night shift week, there were no marked differences in sleep quality. Both morning and afternoon sleep were characterized by deep sleep, and short sleep onset latencies.     

Vigilance on the civil flight deck: incidence of sleepiness and sleep during long-haul flights and associated changes in physiological parameters

The study investigated sleepiness and sleep in aircrew during long-haul flights. The objectives were to identify loss of alertness and to recommend a practical approach to the design of an alerting system to be used by aircrew to prevent involuntary sleep. The flights were between London and Miami, covering both day- and night-time sectors, each with a duration of ~9 h. The subjects were 12 British Airways pilots. Various physiological variables were measured that could potentially be used to indicate the presence of drowsiness and involuntary sleep: brain electrical activity (electroencephalogram, EEG), eye movements via the electro-oculogram (EOG), wrist activity, head movements and galvanic skin resistance. The EEG and EOG identified sleepiness and sleep, as well as being potential measures on which to base an alarm system. Ten pilots either slept or showed evidence of sleepiness as assessed by the EEG and EOG. Many of the episodes of sleepiness lasted < 20 s, which could mean that the subjects were unaware of their occurrence and of the potential consequences on performance and vigilance. All physiological parameters showed changes during sleep, although only the EEG and EOG were modified by sleepiness. During sleep, skin resistance was increased, and wrist activity and head movements were absent for long periods. The study indicated that the measurement of eye movements (either alone or in combination with the EEG), wrist activity or head movement may be used as the basis of an alarm system to prevent involuntary sleep. Skin resistance is considered to be unsuitable, however, being related in a more general way to fatigue rather than to sleep episodes. The optimal way to monitor the onset of sleep would be to measure eye movements; however, this is not feasible in the flight deck environment at the present time due to the intrusive nature of the recording methodology. Wrist activity is therefore recommended as the basis of an alertness alarm. Such a device would alert the pilot after ~4–5 min of wrist inactivity, since this duration has been shown by the present study to be associated with sleep. The possibility that sleep inertia (reduced alertness immediately after awakening from sleep) could follow periods of sleep lasting 5 min needs to be considered. The findings reported here might be applicable to other occupational environments where fatigue and sleepiness are known to occur.     

Vigilance on the civil flight deck: incidence of sleepiness and sleep during long-haul flights and associated changes in physiological parameters

The study investigated sleepiness and sleep in aircrew during long-haul flights. The objectives were to identify loss of alertness and to recommend a practical approach to the design of an alerting system to be used by aircrew to prevent involuntary sleep. The flights were between London and Miami, covering both day- and night-time sectors, each with a duration of approximately 9 h. The subjects were 12 British Airways pilots. Various physiological variables were measured that could potentially be used to indicate the presence of drowsiness and involuntary sleep: brain electrical activity (electroencephalogram, EEG), eye movements via the electro-oculogram (EOG), wrist activity, head movements and galvanic skin resistance. The EEG and EOG identified sleepiness and sleep, as well as being potential measures on which to base an alarm system. Ten pilots either slept or showed evidence of sleepiness as assessed by the EEG and EOG. Many of the episodes of sleepiness lasted < 20 s, which could mean that the subjects were unaware of their occurrence and of the potential consequences on performance and vigilance. All physiological parameters showed changes during sleep, although only the EEG and EOG were modified by sleepiness. During sleep, skin resistance was increased, and wrist activity and head movements were absent for long periods. The study indicated that the measurement of eye movements (either alone or in combination with the EEG), wrist activity or head movement may be used as the basis of an alarm system to prevent involuntary sleep. Skin resistance is considered to be unsuitable, however, being related in a more general way to fatigue rather than to sleep episodes. The optimal way to monitor the onset of sleep would be to measure eye movements; however, this is not feasible in the flight deck environment at the present time due to the intrusive nature of the recording methodology. Wrist activity is therefore recommended as the basis of an alertness alarm. Such a device would alert the pilot after approximately 4-5 min of wrist inactivity, since this duration has been shown by the present study to be associated with sleep. The possibility that sleep inertia (reduced alertness immediately after awakening from sleep) could follow periods of sleep lasting 5 min needs to be considered. The findings reported here might be applicable to other occupational environments where fatigue and sleepiness are known to occur.     

The role of naps in alleviating sleepiness during an irregular sleep-wake schedule

Abstract

Four groups of four subjects have been studied in an isolation unit on an irregular schedule of sleep and waking. Sleep periods of 6h duration were taken at irregular times during the 9-day protocol; they were arranged such that the waking periods lasted 6, 12 or 18 h and were also in an irregular sequence. One-hour naps were taken after 6 and 12 h of waking if the next sleep was not due then. Just before either a nap or a full sleep, subjects assessed their sleepiness and, on rising, they assessed how well sleep had been initiated and maintained. A mathematical model was developed which enabled the effects upon sleepiness of circadian rhythmicity, of the length of the waking period since the last full sleep (time-since-sleep), and of the subjective value of the previous full sleep or nap to be independently quantified. Results showed that there was a tendency for sleepiness to increase, in some cases quite markedly, as the irregular schedule proceeded. There were also effects upon sleepiness due to circadian changes and the amount of time since the last full sleep. These two effects were similar in size and larger than the benefits to be derived from a previous nap or full sleep that had been assessed at the time as having been a good one. We conclude that one-hour naps can play a small role in decreasing an individual's sleepiness and suggest possible means by which their subjective value might be increased.     

Effect of sleep loss on core temperature when movement is controlled

Nine subjects were studied for 16 days in an isolation unit where they lived on normal time, working at a decision-making, computer-driven task during the daytime. Interspersed among these control days were three occasions when sleep was curtailed. Rectal temperature and activity (non-dominant wrist) were measured throughout. Any effects of sleep loss on core temperature and activity were assessed by comparing these variables on control days with values during the daytime immediately following sleep loss, and during the next (recovery) day. During the daytime following sleep loss, activity showed no significant changes. By contrast, core temperature was significantly lower, particularly after the night of complete sleep loss. On recovery days also, activity was not significantly changed from control days but core temperatures during work were significantly lower than on control days if there had been no sleep the previous night. These results indicate that the effects of sleep loss on core temperature can persist for at least 24 h, and that they occur in the absence of parallel changes in activity.     

Effect of sleep loss on core temperature when movement is controlled

Nine subjects were studied for 16 days in an isolation unit where they lived on normal time, working at a decision-making, computer-driven task during the daytime. Interspersed among these control days were three occasions when sleep was curtailed. Rectal temperature and activity (non-dominant wrist) were measured throughout. Any effects of sleep loss on core temperature and activity were assessed by comparing these variables on control days with values during the daytime immediately following sleep loss, and during the next (recovery) day. During the daytime following sleep loss, activity showed no significant changes. By contrast, core temperature was significantly lower, particularly after the night of complete sleep loss. On recovery days also, activity was not significantly changed from control days but core temperatures during work were significantly lower than on control days if there had been no sleep the previous night. These results indicate that the effects of sleep loss on core temperature can persist for at least 24 h, and that they occur in the absence of parallel changes in activity.     

Ageing, physical fitness and shiftwork tolerance

The effects of ageing and physical fitness on shiftwork tolerance are reviewed. Ageing is one of the most cited factors decreasing the health of shiftworkers. Although long-term prospective studies on ageing are few, shiftworkers over 40-45 years of age seem to sleep worse after night, but not after morning shifts. Sleepiness after successive night shifts is also decreased by age although older shiftworkers' ability to resist acute sleep loss seems to be even better. The reasons for the altered sleep and wakefulness of older shiftworkers are probably related to changes in circadian rhythms, especially higher 'morningness'. Sleep need may also decrease with age which could explain some of the differences found in sleep length. Physical fitness as a factor increasing tolerance to shiftwork is a recent finding. Although the effects of physical activity on sleep have been studied in detail, the relationship of physical fitness to sleep is still a controversial issue. In shiftworkers, moderate physical training has been shown to increase sleep length and night-time alertness. It has not been shown, however, that exercise would quicken the circadian adjustment to night work. It is recommended that work time arrangements should take account of the older workers' changing personal references. Continuous night work should be voluntary after 40 years of age. Moderate physical exercise a few hours before the main sleep period is recommended.     

基于IEEE 802.16e的睡眠模式的研究

IEEE 802.16e介绍了两种睡眠模式来达到能量的有效利用,分别是睡眠模式Ⅰ和睡眠模式Ⅱ,睡眠模式Ⅰ使用二进制指数增长的睡眠周期大小,睡眠模式Ⅱ使用固定的睡眠周期大小。首先对两种睡眠模式的能量消耗和时延性能进行分析,然后引入马尔可夫判定过程,给出一种睡眠模式的选择算法,以满足对于能量消耗和时延的要求,从而达到一种在不同性能要求下的权衡,仿真结果给出了这种选择算法的性能特点。     

Sleep modes effectiveness in backbone networks with limited configurations

In this work we propose a strategy to minimize the impact of energy saving techniques on the performance of an Internet Service Provider network. We study the problem of putting in sleep mode links of a backbone network, while limiting the number of times each device changes its power state (full power mode or sleep mode). Our aim is to limit the number of network configurations, i.e., the change of the current set of network links at full power. We tackle the problem both analytically and by simulations. We first present a model, based on random graph theory, to compute the links energy saving given a traffic variation, QoS requirements, and the number of allowed network configurations. The same analysis is then repeated over two realistic case studies, and with realistic algorithms to choose the set of links in sleep mode. Results show (both analytically and by simulation) that the energy savings with few configurations (two or three per day) are close to the maximum one, in which a new configuration is applied for each traffic matrix. Moreover, we show that few configurations per day limit the number of overhead messages required for exchanging information about the device state. Thus, we can conclude that a practical implementation of sleep mode strategies for network operators is to define, on the basis of typical traffic trend, few configurations to be activated in specific time instants.     

Work/rest cycles in railroad operations: effects of shorter than 24-h shift work schedules and on-call schedules on sleep

The current study examined the frequency with which shorter than 24-h work/rest cycles occur in locomotive engineer work schedules, and what effects these work/rest cycles had on sleep quantity and sleep quality. The results indicated that shorter than 24-h work/rest cycles occurred in 33.6% of the work days reported by 198 locomotive engineers. In addition, the shorter than 24-h work/rest cycles occurred more frequently in work schedules that created an on-call work system, such as road pool turn and extra board assignments, than in work schedules that used more predictable or regular work times, such as regular road assignments and yard/local work. As would be expected, when engineers worked shorter than 24-h work/rest cycles, they reported less sleep and poorer sleep than under the longer than 24-h work/rest cycles. Similarly, on-call work assignments resulted in less sleep and poorer sleep than regular work assignments. These results indicate that specific aspects of the work schedules used in railroad operations, particularly on-call operations that result in shorter than 24-h work/rest cycles, can lead to increased sleep-related problems. Although the North American railroad industry is making significant changes in on-call operations to minimize sleep-related problems from on-call schedules, better fatigue-related models validated within the railroad industry are needed.     

Work/rest cycles in railroad operations: effects of shorter than 24-h shift work schedules and on-call schedules on sleep

The current study examined the frequency with which shorter than 24-h work/rest cycles occur in locomotive engineer work schedules, and what effects these work/rest cycles had on sleep quantity and sleep quality. The results indicated that shorter than 24-h work/rest cycles occurred in 33.6% of the work days reported by 198 locomotive engineers. In addition, the shorter than 24-h work/rest cycles occurred more frequently in work schedules that created an on-call work system, such as road pool turn and extra board assignments, than in work schedules that used more predictable or regular work times, such as regular road assignments and yard/local work. As would be expected, when engineers worked shorter than 24-h work/rest cycles, they reported less sleep and poorer sleep than under the longer than 24-h work/rest cycles. Similarly, on-call work assignments resulted in less sleep and poorer sleep than regular work assignments. These results indicate that specific aspects of the work schedules used in railroad operations, particularly on-call operations that result in shorter than 24-h work/rest cycles, can lead to increased sleep-related problems. Although the North American railroad industry is making significant changes in on-call operations to minimize sleep-related problems from on-call schedules, better fatigue-related models validated within the railroad industry are needed.     

基于心率变异性分析的睡眠音乐推荐系统

现有睡眠监测研究主要是针对睡眠质量提出非干扰式监测方法的研究,而缺乏对睡眠质量主动调节方法的研究。基于心率变异性(HRV)分析的精神状态以及睡眠分期研究主要集中在这两种信息的获取上,而这两种信息的获取需要佩戴专业医疗设备,并且这些研究缺乏对信息的应用以及调整。音乐可以作为一种解决睡眠问题的非药物类方法,但现有音乐推荐方法并未考虑个体睡眠及精神状态的差异。针对以上问题提出了一种基于移动设备的精神压力和睡眠状态的音乐推荐系统。首先,用手表采集光体积扫描计信号来提取特征并计算心率;其次,将采集的信号通过蓝牙传递给手机,手机通过这些信号评估人的精神压力以及睡眠状态来播放调整音乐;最后,根据个体每晚的入眠时间进行音乐推荐。实验结果表明,在使用睡眠音乐推荐系统后,用户睡眠总时长相较于使用前增长11.0%。     

The effect of four-day round trip flights over 10 time zones on the sleep-wakefulness patterns of airline flight attendants

To study the effect of a four-day-round trip flight on the sleep-wakefulness of airline flight attendants, subjective sleep-wakefulness and autonomic sleep phases were measured. Forty flight attendants (mean age 33 years, range 21-50) kept daily logs on sleepiness, the time when going to bed, and sleep quality. In addition, the autonomic sleep phases of 21 subjects were studied by the static charge sensitive bed (SCSB) method. After the westward flight, the subjects went to bed approximately 1 -3 h local time earlier during the first few days and were very sleepy compared to the week before the flight. There was a significant increase in the number of awakenings and in the feeling of ‘not being at all rested’ in the mornings. After the return flight eastwards, the subjects were very sleepy on the first evening but slept rather well for about 11 h. During the three following days, sleep restlessness, difficulties in falling asleep, and the feelings of sleepiness in the mornings increased compared to the week before the flights. Four days after the return flight, sleep length and the quality of sleep were, on average, the same as before the flights. According to the SCSB method, there were only small changes in the autonomic sleep phases due to the flights. After the westward flight, quiet sleep increased and intermediate sleep decreased compared to the sleep before the flight. The results indicate that most flight attendants have significant disturbances in sleep quality after transmedian flights. Sleep disturbances increase after both westward and eastward transmedian flights, but differ from each other in specific features.     

Sleep and circadian rhythms in mining operators: limited evidence of adaptation to night shifts

Cumulative sleep deprivation is often associated with work patterns involving night shift or early morning shifts. Adaptation of the circadian system to the shift pattern is reported to promote improved duration and quality of sleep and a concurrent improvement in performance. The current study followed twenty-nine operators at a live-in mining operation working to a seven-day, seven-night shift pattern who collected saliva samples for melatonin measurement, recorded sleep using activity monitors and diaries, and underwent performance testing (psychomotor vigilance task) for one complete roster cycle. The time of onset of melatonin secretion changed significantly (P=0.022) across the week of both Day and Night shifts (2104 h ± 16 min versus 2130 h ± 16 min, respectively), but the small magnitude of the change indicates a lack of true circadian rhythm adaptation to the lifestyle. Total sleep time was longer following the seventh Day shift (associated with a period of 24 h off prior to the commencement of Night shifts). There were no other changes in total sleep time. Further, there were no improvements in sleep onset latency or sleep efficiency on Day or Night shifts. However, reaction times recorded at the end of the shifts slowed across the seven Day and seven Night shifts indicative of impairments in psychomotor performance (F(6,168)=6.087, P<0.001). The results suggest that previous reports of adaptation to consecutive night shifts cannot necessarily be applied to onshore or Australian environments. Adaptation is dependent on factors such as light exposure, environmental conditions, shift parameters such as wake-up, work start and work end times and individual characteristics.     

Tolerance to Shift of Sleep,as Related to the Individual's Circadian Phase Position

The differential adaptation of ‘morning’ and ‘evening’ types to changes in sleep schedules was investigated in two studies. The first was carried out on six female subjects, classified as morning or evening types on the basis of their rhythms of body temperature, and on their answers to a morningness-eveningness questionnaire. Bed times on single nights were varied between 2100h and 0300 h. On each night measurements were taken of latency of sleep onset and of sleep duration. Body temperature, heart rate and respiratory rate were also recorded, as was self-estimated vigilance during the day following the experimental sleep. Evidence of sleep deficiency after late bed limes was found in the morning but not in the evening types; the latter, however, had longer sleep latencies after early bed times than the morning types.

In the second study eight male subjects underwent the same basic experimental treatment, but in this case sleep was monitored for three consecutive nights with a given bed time, in order to examine adaptive processes. The overall results were similar to those of the first study. However, it was also found that, whereas the duration of sleep tended to increase over the three consecutive nights with later bed times in all subjects, this was more marked in the evening than in the morning types.