Identification of major depression patients using machine learning models based on heart rate variability during sleep stages for pre-hospital screening.

With the COVID-19 pandemic causing challenges in hospital admissions globally, the role of home health monitoring in aiding the diagnosis of mental health disorders has become increasingly important. This paper proposes an interpretable machine learning solution to optimise initial screening for major depressive disorder (MDD) in both male and female patients. The data is from the Stanford Technical Analysis and Sleep Genome Study (STAGES). We analyzed 5-min short-term electrocardiogram (ECG) signals during nighttime sleep stages of 40 MDD patients and 40 healthy controls, with a 1:1 gender ratio. After preprocessing, we calculated the time-frequency parameters of heart rate variability (HRV) based on the ECG signals and used common machine learning algorithms for classification, along with feature importance analysis for global decision analysis. Ultimately, the Bayesian optimised extremely randomized trees classifier (BO-ERTC) showed the best performance on this dataset (accuracy 86.32%, specificity 86.49%, sensitivity 85.85%, F1-score 0.86). By using feature importance analysis on the cases confirmed by BO-ERTC, we found that gender is one of the most important factors affecting the prediction of the model, which should not be overlooked in our assisted diagnosis. This method can be embedded in portable ECG monitoring systems and is consistent with the literature results.

Event-related potentials in insomnia reflect altered perception of sleep.

STUDY OBJECTIVES: Insomnia is defined by the subjective complaint of poor sleep as well as daytime impairments. Since polysomnography (PSG) typically shows only modest sleep impairment, some still unidentified property of sleep, not mirrored in PSG, may be modified in insomnia.One possible mechanistic hypothesis is that insomnia patients may be more sensitive to inevitably occurring internal or external stimuli during the night, causing brief sleep disruptions then perceived as wake time. METHODS: Auditory event-related potentials (ERP) to low intensity (50 dB SPL) synthesized guitar tones played continuously throughout two nights of polysomnographically registered sleep were obtained in fifty patients with insomnia disorder (ID, without comorbidities) and 50 age- and sex-matched good sleeper controls (GSC) for each sleep stage and NREM/REM cycle. Phasic and tonic REM were treated as separate stages. Latencies and amplitudes of components P1, N1 and P2 were measured and analyzed by multivariate repeated-measures ANCOVA including effects of group, night, cycle, and age. RESULTS: ID showed reduced P2 amplitudes relative to GSC specifically in phasic REM sleep. The same reduction also correlated with the amount of sleep misperception across groups. Independent component analysis showed a frontal negativity to contribute most to this group difference. CONCLUSIONS: The present finding can be interpreted as increased mismatch negativity (MMN) in ID, reflecting automated detection of change in the auditory system and a concomitant orienting response. Specifically phasic REM sleep appears to be vulnerable to sensory afferences in ID patients, possibly contributing to the perception of being awake. CLINICAL TRIAL INFORMATION: Short name "PERSLEEP 2," URL https://www.drks.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00008965, Registration DRKS00008965.

The effect of COVID-19 lockdowns on sleep time perception: Comparing actigraphy and sleep diary measures.

COVID-19 has become a long-term problem, and global pandemic conditions may persist for years. Researchers are providing mounting evidence of relationships between COVID-19 lockdowns and sleep problems. However, few studies have investigated the impact of home isolation on sleep time perception, especially in comparable social isolation situations with similar pressures. Subjective sleep time perception parameters were derived from sleep diaries. Objective parameters were derived from actigraphy. Subjective and objective data were obtained between February 17 and February 27, 2020 from 70 adult participants subject to COVID-19 related lockdown provisions in China. We divided participants into a home stayers (HS) group (subject to full stay-at home orders) and an area-restricted workers (ARW) group (permitted to work at their nearby workplaces). The HS group demonstrated significantly delayed actigraphy-defined sleep onset time compared to self-reported sleep onset time; this effect was absent in the ARW group. Between-group differences in actigraphy-defined sleep onset time and significant between-group differences for actigraphy-defined and self-reported wake-up time were observed. HS group participants also presented significantly delayed actigraphy-defined wake-up time compared with self-reported wake-up time. No significant effect was found on total sleep time perception. Moreover, sleep/wake time misperception were found to be associated with daylight exposure and physical activity levels respectively. To the extent they are generalizable, these results suggest that lockdown restrictions can affect sleep onset and wake-up time perception but not total sleep time perception. Public health policy should consider such effects in the present pandemic situation and in future emergent public health situations.

Decrease in sleep quality during COVID-19 outbreak.

PURPOSE: The COVID-19 outbreak witnessed in the first months of 2020 has led to unprecedented changes in society's lifestyles. In the current study, we aimed to investigate the effect of this unexpected context on sleep. METHODS: During the COVID-19 outbreak, we performed an online survey with individuals formerly recruited for validation of the Spanish version of the sleep questionnaire Satisfaction, Alertness, Timing, Efficiency, and Duration (SATED). In the current survey, we asked the participants to complete the previously answered questionnaires including the Pittsburgh Sleep Quality Index (PSQI), a modified version of the Epworth Sleepiness Scale (ESS), and the SATED questionnaire. We also assessed the mood by the Profile of Mood States (POMS) questionnaire. RESULTS: The 71 participants were mostly women (75%) with a mean (± SD) age of 40.7 ± 11.9 years. Comparing the previous PSQI score to that during the COVID-19 outbreak, we observed worsening sleep quality (5.45 ± 3.14 to 6.18 ± 3.03 points, p = 0.035). In parallel, there was an increase in the negative mood (p = 0.002). Accordingly, the decrease in sleep quality was substantially correlated with negative mood (p < 0.001). There were no differences in the ESS or SATED. CONCLUSIONS: The COVID-19 outbreak-associated events correlate with decreased sleep quality in association with an increase in negative mood. Considering the importance of sleep for a healthy life, and in particular for immune function, efforts should be made to improve awareness on this matter and to offer psychological assistance to affected individuals.