Figure 1 - uploaded by Monica R Kelly
Content may be subject to copyright.
The Z-PLUS algorithm further differentiates epochs identified as Sleep by Z-ALG into Light Sleep, Deep Sleep, and REM. Abbreviations: eeg, electroencephalography; reM, rapid eye Movement.
Source publication
Background: We previously published the performance evaluation of an automated electroencephalography (EEG)-based single-channel sleep–wake detection algorithm called Z-ALG used by the Zmachine® sleep monitoring system. The objective of this paper is to evaluate the performance of a new algorithm called Z-PLUS, which further differentiates sleep as...
Context in source publication
Context 1
... Z-PLUS algorithm evaluated in this paper further dif- ferentiates epochs identified as sleep by Z-ALG into Light Sleep, Deep Sleep, and REM as shown in Figure 1. The performance of Z-PLUS was evaluated against the same 99-subject data set as used to evaluate the wake-sleep detec- tion performance of Z-ALG. 13 Both Z-PLUS and Z-ALG use EEG data acquired from the contralateral area of the mastoid process located behind the ears, termed A 1 -A 2 , with a signal common (COM) located on the back of the neck or shoulder, as shown in Figure 2. A schematic representation of Z-PLUS is depicted in Figure 3
...
Similar publications
Purpose:
Chronic primary insomnia (CPI) is the most prevalent sleep disorder worldwide. CPI manifests as difficulties in sleep onset, maintaining sleep, prolonged sleep latency, and daytime impairment and is often accompanied by cognitive problems such as poor academic performance, poor attention, and decreased memory. The most popular explanation...
We conducted this study to assess the usefulness of polysomnography (PSG) in patients referred to the sleep disorders center (SDC) with chronic insomnia. Sixty-seven patients with chronic insomnia who underwent overnight polysomnography were included in the study. Clinical diagnoses were reported using the second revision of the International Class...
Citations
... Automated sleep staging methods are preferred over manual staging techniques due to their enormous advantages namely in computational efficiency in processing large data, producing consistent and reliable results, scalability, and greater statistical power and cost-effectiveness. [7][8][9][10][11][12][13][14][15][16] Sleep is a very essential parameter to assess one's health condition. The deep sleep stage is the most essential of all the sleep stages mentioned by AASM. ...
Electroencephalography (EEG) is a neuro signal reflecting brain activity. These signals provide information about brain activity, eye movements, and muscle tone, which can be used to determine the sleep stage. Categorizing sleep stages can be done manually by visually. Alternatively, automated algorithms can be developed using machine learning techniques to classify sleep stages based on signal features and patterns. This paper aims to automatically classify sleep stages based on extracted patterns from EEG signals. A fuzzy min-max neural network is proposed and implemented for sleep stage classification and clustering. The paper concludes that the fuzzy min-max neural network outperforms other tested methods in sleep stage classification. The models implemented in the study include K-Nearest Neighbor (KNN), Random Forest, Decision Tree, XGBoost, AdaBoost, Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Convolutional Neural Network (CNN), and the fuzzy min-max classifier. The results indicate that the fuzzy classifier achieves the highest accuracy of 86%, followed by the CNN model with 81%. Among the machine learning algorithms, Random Forest with an accuracy of 55.46%, followed by XGBoost with 53.18%, surpassing the other algorithms used in the experiment. AdaBoost and Gaussian Naive Bayes both achieve an accuracy of 45.10%. Decision Tree, KNN, LDA, and QDA yield accuracies of 37.66%, 16.46%, 28.57%, and 29.5% respectively. These findings demonstrate the efficiency of the fuzzy min-max neural network and the superiority of the fuzzy classifier and CNN models in sleep stage classification, indicating their potential for accurate automated sleep stage analysis.
... Since subjects turned on the HST right before going to sleep, SOL was found using the difference in minutes between sleep onset and the start of the nightly HST recording. SOL itself is often calculated using the rst epoch of stage 2 sleep 31 , but Z-ALG returns stages 1 and 2 together as light sleep 27 , so we cannot distinguish between stage 1 and stage 2. To overcome this, we found the rst hour where the individual was asleep for more than 10 minutes. Within this rst hour of substantial sleep, we found the exact time when sleep started. ...
Body temperature should be tightly regulated for optimal sleep. However, various extrinsic and intrinsic factors can alter body temperature during sleep. We examined how sleep and cardiovascular health metrics were affected by sleeping for one week with vs. without an active temperature-controlled mattress cover (the Eight Sleep Pod). 54 subjects wore a home sleep test device (HST) for four nights each with the Pod on and off (> 300 total HST nights), while also collecting sleeping heart rate (HR) and heart rate variability (HRV). With the Pod on, men’s deep and light sleep increased by 14 and 23 min, respectively, while women’s REM sleep increased by 9 min. Sleeping HR (-2%) and HRV (+ 7%) significantly improved with Pod on. To our knowledge, this is the first study to show that a continuously temperature-regulated bed surface can significantly 1) modify time spent in sleep stages at specific points in the night, 2) improve thermal comfort and perceived sleep quality, and 3) enhance cardiovascular recovery (HR and HRV) during sleep.
... For the sleep protocol, participants were instructed to wear the Zmachine (General Sleep, Columbus, OH, USA), a single channel electroencephalogram (EEG) that monitors sleep duration and stages, during sleep on both nights. The Zmachine has been validated against polysomnography (PSG) [48] and is widely used in sleep research studies [22,26,30,49]. Participants were instructed to put the Zmachine on 30 min before going to bed. ...
(1) Background: Sleep may be a factor that influences the taste–dietary intake relationship. The effect of sleep on salt taste measures has not been adequately studied, and no standardized methodology has been developed for measuring salt taste preference. (2) Methods: A sweet taste forced-choice paired-comparison test was adapted and validated to determine salt taste preference. In a randomized cross-over trial, participants slept a curtailed night (33% reduction in sleep duration) and a habitual night, confirmed by a single-channel electroencephalograph. Salt taste tests were conducted the day after each sleep condition using five aqueous NaCl solutions. One 24-h dietary recall was obtained after each taste test. (3) Results: The adapted forced-choice paired-comparison tracking test reliably determined salt taste preference. No changes in salt taste function (intensity slopes: p = 0.844) or hedonic measures (liking slopes: p = 0.074; preferred NaCl concentrations: p = 0.092) were observed after the curtailed sleep condition compared to habitual sleep. However, sleep curtailment disrupted the association between liking slope and energy-corrected Na intake (p < 0.001). (4) Conclusions: The present study serves as the first step toward more standardized taste assessments to facilitate comparison between studies and suggests accounting for sleep when exploring taste–diet relationships.
... During the past decade, there have been various advances in automated sleep staging of PSGs data using the ability of deep learning methods to automatically extract features from data that are relevant to the classification, moreover, the performance of deep neural networks continues to improve as datasets become larger. [46][47][48][49][50] To analyse the effects of electroacupuncture combined with CBT-I on sleep architecture, deep learning will be used to analyse PSGs in various clinical settings in this study. But due to lack of acceptability among patients, some of the participants received placebo treatment will drop, which may make analyses no meaningful due to the small number of participants in each arm. ...
Introduction
Insomnia affects physical and mental health due to the lack of continuous and complete sleep architecture. Polysomnograms (PSGs) are used to record electrical information to perform sleep architecture using deep learning. Although acupuncture combined with cognitive–behavioural therapy for insomnia (CBT-I) could not only improve sleep quality, solve anxiety, depression but also ameliorate poor sleep habits and detrimental cognition. Therefore, this study will focus on the effects of electroacupuncture combined with CBT-I on sleep architecture with deep learning.
Methods and analysis
This randomised controlled trial will evaluate the efficacy and effectiveness of electroacupuncture combined with CBT-I in patients with insomnia. Participants will be randomised to receive either electroacupuncture combined with CBT-I or sham acupuncture combined with CBT-I and followed up for 4 weeks. The primary outcome is sleep quality, which is evaluated by the Pittsburgh Sleep Quality Index. The secondary outcome measures include a measurement of depression severity, anxiety, maladaptive cognitions associated with sleep and adverse events. Sleep architecture will be assessed using deep learning on PSGs.
Ethics and dissemination
This trial has been approved by the institutional review boards and ethics committees of the First Affiliated Hospital of Sun Yat-sun University (2021763). The results will be disseminated through peer-reviewed journals. The results of this trial will be disseminated through peer-reviewed publications and conference abstracts or posters.
Trial registration number
CTR2100052502.
... Proposed models have evolved from simple binary detection of sleep-wake to a more fine-grained sleep classification. Examples are automatic frameworks utilizing a single lead EEG, either isolated from PSG recordings or collected using the Zmachine R sleep monitoring system (Wang et al., 2015;Yoon et al., 2017b;Olesen et al., 2021), or a single lead ECG either isolated from PSG studies or acquired using a variety of strategies (Li et al., 2018;Walch et al., 2019;Sridhar et al., 2020) potentially complemented by respiratory effort (Fonseca et al., 2015;Bakker et al., 2021). The model validated in this study provides a significant contribution toward the methodological advancement of automatic sleep stage scoring relying solely on a single lead ECG signal, acquired via a wearable patch. ...
Background
The rapid advancement in wearable solutions to monitor and score sleep staging has enabled monitoring outside of the conventional clinical settings. However, most of the devices and algorithms lack extensive and independent validation, a fundamental step to ensure robustness, stability, and replicability of the results beyond the training and testing phases. These systems are thought not to be feasible and reliable alternatives to the gold standard, polysomnography (PSG).
Materials and methods
This validation study highlights the accuracy and precision of the proposed heart rate (HR)-based deep-learning algorithm for sleep staging. The illustrated solution can perform classification at 2-levels (Wake; Sleep), 3-levels (Wake; NREM; REM) or 4- levels (Wake; Light; Deep; REM) in 30-s epochs. The algorithm was validated using an open-source dataset of PSG recordings (Physionet CinC dataset, n = 994 participants, 994 recordings) and a proprietary dataset of ECG recordings (Z3Pulse, n = 52 participants, 112 recordings) collected with a chest-worn, wireless sensor and simultaneous PSG collection using SOMNOtouch.
Results
We evaluated the performance of the models in both datasets in terms of Accuracy (A), Cohen’s kappa (K), Sensitivity (SE), Specificity (SP), Positive Predictive Value (PPV), and Negative Predicted Value (NPV). In the CinC dataset, the highest value of accuracy was achieved by the 2-levels model (0.8797), while the 3-levels model obtained the best value of K (0.6025). The 4-levels model obtained the lowest SE (0.3812) and the highest SP (0.9744) for the classification of Deep sleep segments. AHI and biological sex did not affect scoring, while a significant decrease of performance by age was reported across the models. In the Z3Pulse dataset, the highest value of accuracy was achieved by the 2-levels model (0.8812), whereas the 3-levels model obtained the best value of K (0.611). For classification of the sleep states, the lowest SE (0.6163) and the highest SP (0.9606) were obtained for the classification of Deep sleep segment.
Conclusion
The results of the validation procedure demonstrated the feasibility of accurate HR-based sleep staging. The combination of the proposed sleep staging algorithm with an inexpensive HR device, provides a cost-effective and non-invasive solution deployable in the home environment and robust across age, sex, and AHI scores.
... It is possible for the statistic to be negative, which implies that there is no effective agreement between the two raters or the agreement is worse than random. It is well-known that there exists inevitable variability and disagreements between human experts on epoch ratings -Wang et al. [89] found kappas of 0.72-0.85 between two human raters and kappas of 0.82-0.85 between well-performing raters. Therefore, a Cohen's kappa of 0.8 is considered to show strong agreement once we take into account such variabilities. ...
Clinical-grade wearable sleep monitoring is a challenging problem since it requires concurrently monitoring brain activity, eye movement, muscle activity, cardio-respiratory features, and gross body movements. This requires multiple sensors to be worn at different locations as well as uncomfortable adhesives and discrete electronic components to be placed on the head. As a result, existing wearables either compromise comfort or compromise accuracy in tracking sleep variables. We propose PhyMask, an all-textile sleep monitoring solution that is practical and comfortable for continuous use and that acquires all signals of interest to sleep solely using comfortable textile sensors placed on the head. We show that PhyMask can be used to accurately measure all the signals required for precise sleep stage tracking and to extract advanced sleep markers such as spindles and K-complexes robustly in the real-world setting. We validate PhyMask against polysomnography and show that it significantly outperforms two commercially-available sleep tracking wearables – Fitbit and Oura Ring.
... A great number of scientific literatures about automatic sleep staging detection were presented. The majority of these scientific literatures use single-channel EEG recordings for automatic sleep staging [19] and in most cases classified models are built on extracted features. Features are extracted from linear or nonlinear. ...
... Res. Public Health 2022, 19, 741 ...
A steady increase in sleep problems has been observed along with the development of society. Overnight exposure to a static magnetic field has been found to improve sleep quality; however, such studies were mainly based on subjective evaluation. Thus, the presented data cannot be used to infer sleep architecture in detail. In this study, the subjects slept on a magneto-static mattress for four nights, and self-reported scales and electroencephalogram (EEG) were used to determine the effect of static magnetic field exposure (SMFE) on sleep. Machine learning operators, i.e., decision tree and supporting vector machine, were trained and optimized with the open access sleep EEG dataset to automatically discriminate the individual sleep stages, determined experimentally. SMEF was found to decrease light sleep duration (N2%) by 3.51%, and sleep onset latency (SOL) by 15.83%, while it increased deep sleep duration (N3%) by 8.43%, compared with the sham SMFE group. Further, the overall sleep efficiency (SE) was also enhanced by SMFE. It is the first study, to the best of our knowledge, where the change in sleep architecture was explored by SMFE. Our findings will be useful in developing a non-invasive sleep-facilitating instrument.
... The algorithm in ZM has been compared to polysomnography (PSG) in adults with and without chronic sleep issues within a laboratory setting [23,24], and we found that ZM can be feasibly applied to children and adults for multiple days of measurements in free-living [12]. The ZM device demonstrated a high degree of validity for detecting sleep versus awake with a sensitivity, specificity, positive predictive value, and negative predictive values of 95.5%, 92.5%, 98%, and 84.2%, respectively [24]. ...
With the emergence of machine learning for the classification of sleep and other human behaviors from accelerometer data, the need for correctly annotated data is higher than ever. We present and evaluate a novel method for the manual annotation of in-bed periods in accelerometer data using the open-source software Audacity®, and we compare the method to the EEG-based sleep monitoring device Zmachine® Insight+ and self-reported sleep diaries. For evaluating the manual annotation method, we calculated the inter- and intra-rater agreement and agreement with Zmachine and sleep diaries using interclass correlation coefficients and Bland–Altman analysis. Our results showed excellent inter- and intra-rater agreement and excellent agreement with Zmachine and sleep diaries. The Bland–Altman limits of agreement were generally around ±30 min for the comparison between the manual annotation and the Zmachine timestamps for the in-bed period. Moreover, the mean bias was minuscule. We conclude that the manual annotation method presented is a viable option for annotating in-bed periods in accelerometer data, which will further qualify datasets without labeling or sleep records.
... Validation studies of Zmachine have been conducted in healthy subjects, patients with insomnia and psychiatric patients. [37][38][39] The detailed methods of Zmachine have been reported. 37 Participants can easily attach Zmachine at night by themselves. ...
... 65 66 At-home sleep monitors may help to address this problem by allowing participants to sleep in their beds while maintaining sufficient agreement with PSG. 39 The effect of lemborexant on objective sleep parameters has been evaluated by PSG, 34 67 but there are no objective data regarding natural sleep conditions except for subjective evaluations such as sleep diaries or questionnaires. Therefore, Zmachine will make it possible to evaluate the effect of lemborexant on objective sleep parameters under near-natural sleep conditions. ...
Introduction
Bidirectional associations have been reported between sleep disturbance and both cognitive impairment, including Alzheimer’s disease and amyloid beta-peptide (Aβ) accumulation. These relationships can be explained by the glymphatic system, which acts as a garbage drainage system in the brain. As interstitial fluid dynamics are suggested to increase during sleep, clearance of Aβ can be influenced by sleep disturbance or deprivation. We hypothesised that using lemborexant, an orexin receptor antagonist, to improve sleep quality would also improve the function of the glymphatic system. We plan to examine the effect of lemborexant on sleep quality and the glymphatic system among patients with insomnia disorder.
Methods and analysis
This pilot study is designed as an open-label, single-arm, single-centre trial. Thirty patients aged 50 years and over with insomnia will be recruited. The participants will take lemborexant (5 mg) at bedtime for 12 weeks and undergo a home-based sleep study at baseline and weeks 4 and 12, as well as MRI examinations to evaluate the glymphatic system at baseline and week 12. The primary outcome will be changes in objective sleep parameters as evaluated using a sleep monitoring system. The secondary outcomes will be changes in subjective sleep parameters. The relationships between changes in sleep parameters and the glymphatic system will be evaluated using diffusion tensor image analysis along the perivascular space, which is called the ALPS-index. Sleep parameters and the ALPS-index will be analysed using a paired t-test or Pearson’s correlation coefficient.
Ethics and dissemination
The study protocol was approved by Nagoya University Certified Review Board. The findings from this research will be published in peer-reviewed journals and be presented at local, national and international conferences.
Trial registration number
jRCTs041210024.
... The Zmachine utilizes single channel electroencephalography (EEG) data from two sensors placed at the mastoids (A 1 and A 2 ) to record the electrical activity of the brain. The EEG-signal is processed into sleep stages by the Zmachine algorithm (Kaplan et al., 2014;Wang et al., 2015). Though, the Zmachine is a promising tool, only one study have examined the feasibility of this device in adolescents (Lunsford-Avery et al., 2020). ...
... Participants were instructed to attach three sensors to the back of the head approximately 30 min prior to bedtime which was defined as "when you are lying in bed and you are ready to close your eyes and go to sleep". One sensor was placed on the neck below the hair line (ground) and one behind each ear on the differential mastoids (signal) (Wang et al., 2015). We custom-made an elastic pocket for the Zmachine, which the participants were instructed to attach to an elastic waist belt at bedtime ( Fig. 2 and Additional file 1). ...
... Kaplan et al. found that the Zmachine algorithm has high sensitivity (95.5%) and specificity (92.5%) when compared to polysomnographic technologists in scoring sleep and wake in adults (Kaplan et al., 2014). Wang et al. compared the Z-PLUS algorithm in conjunction with the Z-ALG algorithm to sleep stages scored by polysomnographic technologists and found that it has high sensitivity ranging between 72 to 91% in adults (Kaplan et al., 2014;Wang et al., 2015). ...
Background
Sleep is a crucial part of our lives and insufficient sleep has been linked to several health disorders in both children and adults. However, most studies are based on single night laboratory polysomnography, actigraphy, or sleep diaries. The primary aim of this study was to evaluate compliance to and perceived feasibility of the Zmachine insight+ for assessment of habitual sleep parameters in a sample of children and adults for six nights. The secondary aim was to report sleep parameters derived from the Zmachine.
Methods
We analyzed data from 12 families who participated in the SCREENS pilot trial (2018–2019). Children (n=14) and adults (n=19) had to undergo three nights of EEG-based sleep assessment at baseline and follow-up. We assessed compliance to the sleep assessment protocol and summarized perceived feasibility in children and adults. Summary estimates were computed for total sleep time, sleep onset latency, wake after sleep onset, light sleep, deep sleep, and rapid eye movement sleep.
Results
Compliance to the sleep assessment protocol was high with 92.9 and 89.4% of children and adults meeting the a priori specified compliance goal of at least two out of three nights of complete sleep data at both baseline and follow-up. In general, the protocol was perceived as feasible, with low prevalence of sleep disruption and only minor issues, e.g. difficulties with removing sensors. Results on sleep parameters indicate large within group variation.
Conclusions
Our findings support the use of a self-administered EEG-based habitual sleep assessment protocol, including multiple days of measurement, in children and adults.
Trial registration
Cilinicaltrials.gov: NCT03788525 [Secondary outcome measures; Retrospectively registered; 27th December, 2018].
