Sleep staging based on polysomnography (PSG) performed by human experts is the de facto “gold standard” for the objective measurement of sleep. PSG and manual sleep staging is, however, personnel-intensive and time-consuming and it is thus impractical to monitor a person’s sleep architecture over extended periods. Here, we present a novel, low-cost, automatized, deep learning alternative to PSG sleep staging that provides a reliable epoch-by-epoch four-class sleep staging approach (Wake, Light [N1 + N2], Deep, REM) based solely on inter-beat-interval (IBI) data. Having trained a multi-resolution convolutional neural network (MCNN) on the IBIs of 8898 full-night manually sleep-staged recordings, we tested the MCNN on sleep classification using the IBIs of two low-cost (<EUR 100) consumer wearables: an optical heart rate sensor (VS) and a breast belt (H10), both produced by POLAR®. The overall classification accuracy reached levels comparable to expert inter-rater reliability for both devices (VS: 81%, $\kappa$ = 0.69; H10: 80.3%, $\kappa$ = 0.69). In addition, we used the H10 and recorded daily ECG data from 49 participants with sleep complaints over the course of a digital CBT-I-based sleep training program implemented in the App NUKKUAA™. As proof of principle, we classified the IBIs extracted from H10 using the MCNN over the course of the training program and captured sleep-related changes. At the end of the program, participants reported significant improvements in subjective sleep quality and sleep onset latency. Similarly, objective sleep onset latency showed a trend toward improvement. Weekly sleep onset latency, wake time during sleep, and total sleep time also correlated significantly with the subjective reports. The combination of state-of-the-art machine learning with suitable wearables allows continuous and accurate monitoring of sleep in naturalistic settings with profound implications for answering basic and clinical research questions.

The purpose of the present study was to evaluate the performance of a low-cost commercial smartwatch, the Xiaomi Mi Band (MB), in extracting physical activity and sleep-related measures and show its potential use in addressing questions that require large-scale real-time data and/or intercultural data including low-income countries. We evaluated physical activity and sleep-related measures and discussed the potential application of such devices for large-scale step and sleep data acquisition. To that end, we conducted two separate studies. In Study 1, we evaluated the performance of MB by comparing it to the GT3X (ActiGraph, wGT3X-BT), a scientific actigraph used in research, as well as subjective sleep reports. In Study 2, we distributed the MB across four countries (Austria, Germany, Cuba, and Ukraine) and investigated physical activity and sleep among these countries. The results of Study 1 indicated that MB step counts correlated highly with the scientific GT3X device, but did display biases. In addition, the MB-derived wake-up and total-sleep-times showed high agreement with subjective reports, but partly deviated from GT3X predictions. Study 2 revealed similar MB step counts across countries, but significant later wake-up and bedtimes for Ukraine than the other countries. We hope that our studies will stimulate future large-scale sensor-based physical activity and sleep research studies, including various cultures.

Summary: Due to insufficient treatment options for insomnia, effective solutions are urgently needed. We evaluated the effects of a CBT‐I‐based app combining sleep training with subjective and objective sleep monitoring on (i) sleep and (ii) subjective‐objective sleep discrepancies (SOSD). Fifty‐seven volunteers (20–76 years; M_Age = 45.67 ± 16.38; 39 female) suffering from sleep problems were randomly assigned to an experimental group (EG, n = 28) or a waitlist control group (CG, n = 29). During the 6‐week app phase, the EG used the CBT‐I‐based programme and a heart rate sensor for daily sleep monitoring and ‐feedback, while the CG used sleep monitoring only. Sleep was measured (i) subjectively via questionnaires (Insomnia Severity Index, ISI; Pittsburgh Sleep Quality Index, PSQI), (ii) objectively via ambulatory polysomnography (PSG), and (iii) continuously via heart‐rate sensor and sleep diaries. Data revealed interactions for ISI (p = 0.003, ƞ²_part = 0.11) and PSQI (p = 0.050, ƞ²_part = 0.05), indicating training‐specific improvements in EG, yet not in CG. While PSG‐derived outcomes appear to be less training‐specific, a tendential reduction in wake after sleep onset (WASO) was found in EG (p = 0.061, d = 0.55). Regarding changes in SOSD, the results indicate improvements during the app phase (EG) for sleep efficiency, sleep onset latency, and WASO (p ≤ 0.022, d ≥ 0.46); for total sleep time both groups showed a SOSD reduction. The findings indicate beneficial effects of a novel smartphone app on sleep and SOSD. More scientific evaluation of such digital programmes is needed to ultimately help in reducing the gap in non‐pharmacological insomnia treatment.

More and more people quantify their sleep using wearables and are becoming obsessed in their pursuit of optimal sleep (“orthosomnia”). However, it is criticized that many of these wearables are giving inaccurate feedback and can even lead to negative daytime consequences. Acknowledging these facts, we here optimize our previously suggested sleep classification procedure in a new sample of 136 self-reported poor sleepers to minimize erroneous classification during ambulatory sleep sensing. Firstly, we introduce an advanced interbeat-interval (IBI) quality control using a random forest method to account for wearable recordings in naturalistic and more noisy settings. We further aim to improve sleep classification by opting for a loss function model instead of the overall epoch-by-epoch accuracy to avoid model biases towards the majority class (i.e., “light sleep”). Using these implementations, we compare the classification performance between the optimized (loss function model) and the accuracy model. We use signals derived from PSG, one-channel ECG, and two consumer wearables: the ECG breast belt Polar® H10 (H10) and the Polar® Verity Sense (VS), an optical Photoplethysmography (PPG) heart-rate sensor. The results reveal a high overall accuracy for the loss function in ECG (86.3 %, $\kappa$ = 0.79), as well as the H10 (84.4%, $\kappa$ = 0.76), and VS (84.2%, $\kappa$ = 0.75) sensors, with improvements in deep sleep and wake. In addition, the new optimized model displays moderate to high correlations and agreement with PSG on primary sleep parameters, while measures of reliability, expressed in intra-class correlations, suggest excellent reliability for most sleep parameters. Finally, it is demonstrated that the new model is still classifying sleep accurately in 4-classes in users taking heart-affecting and/or psychoactive medication, which can be considered a prerequisite in older individuals with or without common disorders. Further improving and validating automatic sleep stage classification algorithms based on signals from affordable wearables may resolve existing scepticism and open the door for such approaches in clinical practice.

A widely endorsed belief is that perceivers imagine their present selves using a different representational format than imagining their future selves (i.e., near future=first-person; distant future=third-person). But is this really the case? Responding to the paucity of work on this topic, here we considered how temporal distance influences the extent to which individuals direct their attention outward or inward during a brief imaginary episode. Using a non-verbal measure of visual perspective taking (i.e., letter-drawing task) our results confirmed the hypothesized relation between temporal distance and conceptions of the self. Whereas simulations of an event in the near future were dominated by a first-person representation of the self, this switched to a third-person depiction when the event was located in the distant future. Critically, this switch in vantage point was restricted to self-related simulations. The theoretical and practical implications of these findings are considered.