THE RELATIONSHIP OF HRV TO SLEEP EEG AND SLEEP RHYTHM

Contactless Camera-Based Sleep Staging: The HealthBed Study

Polysomnography (PSG) remains the gold standard for sleep monitoring but is obtrusive in nature. Advances in camera sensor technology and data analysis techniques enable contactless monitoring of heart rate variability (HRV). In turn, this may allow remote assessment of sleep stages, as different HRV metrics indirectly reflect the expression of sleep stages. We evaluated a camera-based remote photoplethysmography (PPG) setup to perform automated classification of sleep stages in near darkness. Based on the contactless measurement of pulse rate variability, we use a previously developed HRV-based algorithm for 3 and 4-class sleep stage classification. Performance was evaluated on data of 46 healthy participants obtained from simultaneous overnight recording of PSG and camera-based remote PPG. To validate the results and for benchmarking purposes, the same algorithm was used to classify sleep stages based on the corresponding ECG data. Compared to manually scored PSG, the remote PPG-based algorithm achieved moderate agreement on both 3 class (Wake-N1/N2/N3-REM) and 4 class (Wake-N1/N2-N3-REM) classification, with average κ of 0.58 and 0.49 and accuracy of 81% and 68%, respectively. This is in range with other performance metrics reported on sensing technologies for wearable sleep staging, showing the potential of video-based non-contact sleep staging.

Machine learning approaches for predicting sleep arousal response based on heart rate variability, oxygen saturation, and body profiles

Objective

Obstructive sleep apnea is a global health concern, and several tools have been developed to screen its severity. However, most tools focus on respiratory events instead of sleep arousal, which can also affect sleep efficiency. This study employed easy-to-measure parameters-namely heart rate variability, oxygen saturation, and body profiles-to predict arousal occurrence.

Methods

Body profiles and polysomnography recordings were collected from 659 patients. Continuous heart rate variability and oximetry measurements were performed and then labeled based on the presence of sleep arousal. The dataset, comprising five body profiles, mean heart rate, six heart rate variability, and five oximetry variables, was then split into 80% training/validation and 20% testing datasets. Eight machine learning approaches were employed. The model with the highest accuracy, area under the receiver operating characteristic curve, and area under the precision recall curve values in the training/validation dataset was applied to the testing dataset and to determine feature importance.

Results

InceptionTime, which exhibited superior performance in predicting sleep arousal in the training dataset, was used to classify the testing dataset and explore feature importance. In the testing dataset, InceptionTime achieved an accuracy of 76.21%, an area under the receiver operating characteristic curve of 84.33%, and an area under the precision recall curve of 86.28%. The standard deviations of time intervals between successive normal heartbeats and the square roots of the means of the squares of successive differences between normal heartbeats were predominant predictors of arousal occurrence.

Conclusions

The established models can be considered for screening sleep arousal occurrence or integrated in wearable devices for home-based sleep examination.

Automatic Detection of the Cyclic Alternating Pattern of Sleep and Diagnosis of Sleep-Related Pathologies Based on Cardiopulmonary Resonance Indices

The cyclic alternating pattern is the periodic electroencephalogram activity occurring during non-rapid eye movement sleep. It is a marker of sleep instability and is correlated with several sleep-related pathologies. Considering the connection between the human heart and brain, our study explores the feasibility of using cardiopulmonary features to automatically detect the cyclic alternating pattern of sleep and hence diagnose sleep-related pathologies. By statistically analyzing and comparing the cardiopulmonary characteristics of a healthy group and groups with sleep-related diseases, an automatic recognition scheme of the cyclic alternating pattern is proposed based on the cardiopulmonary resonance indices. Using the Hidden Markov and Random Forest, the scheme combines the variation and stability of measurements of the coupling state of the cardiopulmonary system during sleep. In this research, the F1 score of the sleep-wake classification reaches 92.0%. In terms of the cyclic alternating pattern, the average recognition rate of A-phase reaches 84.7% on the CAP Sleep Database of 108 cases of people. The F1 score of disease diagnosis is 87.8% for insomnia and 90.0% for narcolepsy.

Lifestyle Modification Using a Wearable Biometric Ring and Guided Feedback Improve Sleep and Exercise Behaviors: A 12-Month Randomized, Placebo-Controlled Study

Purpose: Wearable biometric monitoring devices (WBMD) show promise as a cutting edge means to improve health and prevent disease through increasing accountability. By regularly providing real-time quantitative data regarding activity, sleep quality, and recovery, users may become more aware of the impact that their lifestyle has on their health. The purpose of this study was to examine the efficacy of a biometric tracking ring on improving sleep quality and increasing physical fitness over a one-year period. Methods: Fifty-six participants received a biometric tracking ring and were placed in one of two groups. One group received a 3-month interactive behavioral modification intervention (INT) that was delivered virtually via a smartphone app with guided text message feedback (GTF). The other received a 3-month non-directive wellness education control (CON). After three months, the INT group was divided into a long-term feedback group (LT-GTF) that continued to receive GTF for another nine months or short-term feedback group (ST-GTF) that stopped receiving GTF. Weight, body composition, and VO2max were assessed at baseline, 3months, and 12months for all participants and additionally at 6 and 9months for the ST-GTF and LT-GTF groups. To establish baseline measurements, sleep and physical activity data were collected daily over a 30-day period. Daily measurements were also conducted throughout the 12-month duration of the study. Results: Over the first 3months, the INT group had significant (p<0.001) improvements in sleep onset latency, daily step count, % time jogging, VO2max, body fat percentage, and heart rate variability (rMSSD HRV) compared to the CON group. Over the next 9months, the LT-GTF group continued to improve significantly (p<0.001) in sleep onset latency, daily step count, % time jogging, VO2max, and rMSSD HRV. The ST-GTF group neither improved nor regressed over the latter 9months except for a small increase in sleep latency. Conclusion: Using a WBMD concomitantly with personalized education, encouragement, and feedback, elicits greater change than using a WBMD alone. Additionally, the improvements achieved from a short duration of personalized coaching are largely maintained with the continued use of a WBMD.

Association between Information and Communication Technology Usage and the Quality of Sleep among School-Aged Children during a School Week

Objective. To determine the association between intensity of information and communication technology (ICT) usage and quality of sleep in school-aged children during a school week. Methods. In all 61 subjects, 10–14 years of age, a quasiexperimental laboratory study where criterions for inclusion were absence of prior medical condition and duration of ICT use. A portable device (Holter monitor) was used to measure heart rate variability (HRV) over a 24-hour period, while activity diary was used to record in 15-minute intervals ICT use and sleep and wake up time. Low and high ICT user groups were formed according to their intensity of ICT use. Statistical analysis was done with two independent samples tests and factorial ANCOVA. Results. The higher ICT users showed a lower sleep time standard deviation of normal to normal interval (SDNN) measures in comparison to the low ICT users. Conclusion. The intensive ICT use was associated with poorer quality of sleep indicated by physiological measures among children and adolescents. Knowing the crucial role of healthy sleep in this age, the results are reason for concern.

Heart Rate Variability, Homeostasis, and Brain Function

Measures of heart rate variability (HRV) are major indices of the sympathovagal balance in cardiovascular research. These measures are thought to reflect complex patterns of brain activation as well and HRV is now emerging as a descriptor thought to provide information on the nervous system organization of homeostatic responses in accordance with the situational requirements. Current models of integration equate HRV to the affective states as parallel outputs of the central autonomic network, with HRV reflecting its organization of affective, physiological, “cognitive,” and behavioral elements into a homeostatic response. Clinical application is in the study of patients with psychiatric disorders, traumatic brain injury, impaired emotion-specific processing, personality, and communication disorders. HRV responses to highly emotional sensory inputs have been identified in subjects in vegetative state and in healthy or brain injured subjects processing complex sensory stimuli. In this respect, HRV measurements can provide additional information on the brain functional setup in the severely brain damaged and would provide researchers with a suitable approach in the absence of conscious behavior or whenever complex experimental conditions and data collection are impracticable, as it is the case, for example, in intensive care units.

Assessment of sleep quality in powernapping

The purpose of this study is to assess the Sleep Quality (SQ) in powernapping. The contributed factors for SQ assessment are time of Sleep Onset (SO), Sleep Length (SL), Sleep Depth (SD), and detection of sleep events (K-complex (KC) and Sleep Spindle (SS)). Data from daytime nap for 10 subjects, 2 days each, including EEG and ECG were recorded. The SD and sleep events were analyzed by applying spectral analysis. The SO time was detected by a combination of signal spectral analysis, Slow Rolling Eye Movement (SREM) detection, Heart Rate Variability (HRV) analysis and EEG segmentation using both Autocorrelation Function (ACF), and Crosscorrelation Function (CCF) methods. The EEG derivation FP1-FP2 filtered in a narrow band and used as an alternative to EOG for SREM detection. The ACF and CCF segmentation methods were also applied for detection of sleep events. The ACF method detects segment boundaries based on single channel analysis, while the CCF includes spatial variation from multiple EEG derivation. The results indicate that SREM detection using EEG is possible and can be used as input together with power spectral analysis to enhance SO detection. Both segmentation methods could detect SO as a segment boundary. Additionally they were able to contribute to detection of KC and SS events. The CCF method was more sensitive to spatial EEG changes and the exact segment boundaries varied slightly between the two methods. The HRV analysis revealed, that low and very low frequency variations in the heart rate was highly correlated with the EEG changes during both SO and variations in SD. Analyzing the relationship between the sleep events and SD showed a negative correlation between the Delta and Sigma activity. Analyzing the subjective measurement (SM) showed that there were a positive correlation between the SL and rated SQ. This preliminary study showed that the factors contributing to the overall SQ during powernapping can be assessed markedly better using a fusion of multiple methods. Future studies will include measures of individual performance before and after powernapping and investigate its relation to the assessed SQ.

Detection of sleep disordered breathing by automated ECG analysis

Sleep related breathing disorders are a highly prevalent disease associated with increased risk of cardiovascular complications like chronic arterial hypertension, myocardial infarction or stroke. Gold standard diagnostics (polysomnography) are complex and expensive; the need for simplified diagnostics is therefore obvious. As the ECG can be easily conducted during the night, the detection of sleep related breathing disorders by ECG analysis provides an easy and cheap approach. Using a combination of well known biosignals processing algorithms, we trained the algorithm on 35 pre-scored overnight recordings. We then applied the algorithm on 35 control recordings, achieving a diagnostic accuracy of 77%. We believe that with further improvements in ECG analysis this algorithm can be used for screening diagnostics of obstructive sleep apnea.