Improving actigraph sleep/wake classification with cardio-respiratory signals

Combining Cardiorespiratory Signals and Video-Based Actigraphy for Classifying Preterm Infant Sleep States

The classification of sleep state in preterm infants, particularly in distinguishing between active sleep (AS) and quiet sleep (QS), has been investigated using cardiorespiratory information such as electrocardiography (ECG) and respiratory signals. However, accurately differentiating between AS and wake remains challenging; therefore, there is a pressing need to include additional information to further enhance the classification performance. To address the challenge, this study explores the effectiveness of incorporating video-based actigraphy analysis alongside cardiorespiratory signals for classifying the sleep states of preterm infants. The study enrolled eight preterm infants, and a total of 91 features were extracted from ECG, respiratory signals, and video-based actigraphy. By employing an extremely randomized trees (ET) algorithm and leave-one-subject-out cross-validation, a kappa score of 0.33 was achieved for the classification of AS, QS, and wake using cardiorespiratory features only. The kappa score significantly improved to 0.39 when incorporating eight video-based actigraphy features. Furthermore, the classification performance of AS and wake also improved, showing a kappa score increase of 0.21. These suggest that combining video-based actigraphy with cardiorespiratory signals can potentially enhance the performance of sleep-state classification in preterm infants. In addition, we highlighted the distinct strengths and limitations of video-based actigraphy and cardiorespiratory data in classifying specific sleep states.

Validation of Fitbit Charge 2 Sleep and Heart Rate Estimates Against Polysomnographic Measures in Shift Workers: Naturalistic Study

BACKGROUND

Multisensor fitness trackers offer the ability to longitudinally estimate sleep quality in a home environment with the potential to outperform traditional actigraphy. To benefit from these new tools for objectively assessing sleep for clinical and research purposes, multisensor wearable devices require careful validation against the gold standard of sleep polysomnography (PSG). Naturalistic studies favor validation.

OBJECTIVE

This study aims to validate the Fitbit Charge 2 against portable home PSG in a shift-work population composed of 59 first responder police officers and paramedics undergoing shift work.

METHODS

A reliable comparison between the two measurements was ensured through the data-driven alignment of a PSG and Fitbit time series that was recorded at night. Epoch-by-epoch analyses and Bland-Altman plots were used to assess sensitivity, specificity, accuracy, the Matthews correlation coefficient, bias, and limits of agreement.

RESULTS

Sleep onset and offset, total sleep time, and the durations of rapid eye movement (REM) sleep and non-rapid-eye movement sleep stages N1+N2 and N3 displayed unbiased estimates with nonnegligible limits of agreement. In contrast, the proprietary Fitbit algorithm overestimated REM sleep latency by 29.4 minutes and wakefulness after sleep onset (WASO) by 37.1 minutes. Epoch-by-epoch analyses indicated better specificity than sensitivity, with higher accuracies for WASO (0.82) and REM sleep (0.86) than those for N1+N2 (0.55) and N3 (0.78) sleep. Fitbit heart rate (HR) displayed a small underestimation of 0.9 beats per minute (bpm) and a limited capability to capture sudden HR changes because of the lower time resolution compared to that of PSG. The underestimation was smaller in N2, N3, and REM sleep (0.6-0.7 bpm) than in N1 sleep (1.2 bpm) and wakefulness (1.9 bpm), indicating a state-specific bias. Finally, Fitbit suggested a distribution of all sleep episode durations that was different from that derived from PSG and showed nonbiological discontinuities, indicating the potential limitations of the staging algorithm.

CONCLUSIONS

We conclude that by following careful data processing processes, the Fitbit Charge 2 can provide reasonably accurate mean values of sleep and HR estimates in shift workers under naturalistic conditions. Nevertheless, the generally wide limits of agreement hamper the precision of quantifying individual sleep episodes. The value of this consumer-grade multisensor wearable in terms of tackling clinical and research questions could be enhanced with open-source algorithms, raw data access, and the ability to blind participants to their own sleep data.

Adaptive Multi-Channel Event Segmentation and Feature Extraction for Monitoring Health Outcomes

OBJECTIVE

To develop a multi-channel device event segmentation and feature extraction algorithm that is robust to changes in data distribution.

METHODS

We introduce an adaptive transfer learning algorithm to classify and segment events from non-stationary multi-channel temporal data. Using a multivariate hidden Markov model (HMM) and Fisher's linear discriminant analysis (FLDA) the algorithm adaptively adjusts to shifts in distribution over time. The proposed algorithm is unsupervised and learns to label events without requiring a priori information about true event states. The procedure is illustrated on experimental data collected from a cohort in a human viral challenge (HVC) study, where certain subjects have disrupted wake and sleep patterns after exposure to an H1N1 influenza pathogen.

RESULTS

Simulations establish that the proposed adaptive algorithm significantly outperforms other event classification methods. When applied to early time points in the HVC data, the algorithm extracts sleep/wake features that are predictive of both infection and infection onset time.

CONCLUSION

The proposed transfer learning event segmentation method is robust to temporal shifts in data distribution and can be used to produce highly discriminative event-labeled features for health monitoring.

SIGNIFICANCE

Our integrated multisensor signal processing and transfer learning method is applicable to many ambulatory monitoring applications.

Wearable Device Heart Rate and Activity Data in an Unsupervised Approach to Personalized Sleep Monitoring: Algorithm Validation

Background

The proliferation of wearable devices that collect activity and heart rate data has facilitated new ways to measure sleeping and waking durations unobtrusively and longitudinally. Most existing sleep/wake identification algorithms are based on activity only and are trained on expensive and laboriously annotated polysomnography (PSG). Heart rate can also be reflective of sleep/wake transitions, which has motivated its investigation herein in an unsupervised algorithm. Moreover, it is necessary to develop a personalized approach to deal with interindividual variance in sleep/wake patterns.

Objective

We aimed to develop an unsupervised personalized sleep/wake identification algorithm using multifaceted data to explore the benefits of incorporating both heart rate and activity level in these types of algorithms and to compare this approach’s output with that of an existing commercial wearable device’s algorithms.

Methods

In this study, a total of 14 community-dwelling older adults wore wearable devices (Fitbit Alta; Fitbit Inc) 24 hours a day and 7 days a week over period of 3 months during which their heart rate and activity data were collected. After preprocessing the data, a model was developed to distinguish sleep/wake states based on each individual’s data. We proposed the use of hidden Markov models and compared different modeling schemes. With the best model selected, sleep/wake patterns were characterized by estimated parameters in hidden Markov models, and sleep/wake states were identified.

Results

When applying our proposed algorithm on a daily basis, we found there were significant differences in estimated parameters between weekday models and weekend models for some participants.

Conclusions

Our unsupervised approach can be effectively implemented based on an individual’s multifaceted sleep-related data from a commercial wearable device. A personalized model is shown to be necessary given the interindividual variability in estimated parameters.

[Daytime Sleepiness in Patients with Restless Legs Syndrome: A Risk Factor for Traffic Accidents?]

Daytime Sleepiness in Patients with Restless Legs Syndrome: A Risk Factor for Traffic Accidents? Abstract. Restless legs syndrome (RLS) is a symptom complex of predominantly leg-focused paraesthesias and the associated increased urge to move. Since evening exacerbations are typical, many patients suffer from sleep disorders, which can lead to increased daytime fatigue in the long term. The present retrospective data analysis investigated a potentially relevant relationship between RLS and an increased incidence of traffic accidents due to daytime sleepiness in Swiss road traffic. A direct correlation between RLS and the occurrence of traffic accidents could not be found. Nevertheless, the question of increased daytime sleepiness should not be absent from any (traffic) medical discussion.

Advanced and Accurate Mobile Health Tracking Devices Record New Cardiac Vital Signs

Cardiovascular disease remains the leading cause of death and disease worldwide. As demands on an already resource-constrained healthcare system intensify, disease prevention in the future will likely depend on out-of-office monitoring of cardiovascular risk factors. Mobile health tracking devices that can track blood pressure and heart rate, in addition to new cardiac vital signs, such as physical activity level and pulse wave velocity (PWV), offer a promising solution. An initial barrier is the development of accurate and easily-scalable platforms. In this study, we made a customized smartphone app and used mobile health devices to track PWV, blood pressure, heart rate, physical activity, sleep duration, and multiple lifestyle risk factors in ≈250 adults for 17 continual weeks. Eligible participants were identified by a company database and then were consented and enrolled using only a smartphone app, without any special training given. Study participants reported high overall satisfaction, and 73% of participants were able to measure blood pressure and PWV, <1 hour apart, for at least 14 of 17 weeks. The study population's blood pressure, PWV, heart rate, activity levels, sleep duration, and the interrelationships among these measurements were found to closely match either population averages or values obtained from studies performed in a controlled setting. As a proof-of-concept, we demonstrated the accuracy and ease, as well as many challenges, of using mHealth technology to accurately track PWV and new cardiovascular vital signs at home.

Comparative assessment of sleep quality estimates using home monitoring technology

Poor sleep quality is associated with chronic diseases, weight increase and cognitive dysfunction. Home monitoring solutions offer the possibility of offering tailored sleep coaching interventions. There are several new commercially available devices for tracking sleep, and although they have been tested in sleep laboratories, little is known about the errors associated with the use in the home. To address this issue we performed a study in which we compared the sleep monitoring data from two commercially available systems: Fitbit One and Beddit Pro. We studied 23 subjects using both systems over a week each and analyzed the degree of agreement for different aspects of sleep. The results suggest the need for individual-tailoring of the estimation process. Not only do these models address improved accuracy of sleep quality estimates, but they also provide a framework for the representation and harmonization for monitoring data across studies.

Challenges and Emerging Technologies within the Field of Pediatric Actigraphy

Actigraphy as an objective measure of sleep and wakefulness in infants and children has gained popularity over the last 20 years. However, the field lacks published guidelines for sleep-wake identification within pediatric age groups. The scoring rules vary greatly and although sensitivity (sleep agreement with polysomnography) is usually high, a significant limitation remains in relation to specificity (wake agreement). Furthermore, accurate algorithm output and sleep-wake summaries usually require prior entry from daily logs of sleep-wake periods and artifact-related information (e.g., non-wear time), involving significant parent co-operation. Scoring criteria for daytime naps remains an unexplored area. Many of the problems facing accuracy of measurement are inherent within the field of actigraphy itself, particularly where sleep periods containing significant movements are erroneously classified as wake, and within quiet wakefulness when no movements are detected, erroneously classified as sleep. We discuss the challenges of actigraphy for pediatric sleep, briefly describe the technical basis and consider a number of technological approaches that may facilitate improved classification of errors in sleep-wake discrimination.

Adaptive Heartbeat Modeling for Beat-to-Beat Heart Rate Measurement in Ballistocardiograms

We present a method for measuring beat-to-beat heart rate from ballistocardiograms acquired with force sensors. First, a model for the heartbeat shape is adaptively inferred from the signal using hierarchical clustering. Then, beat-to-beat intervals are detected by finding positions where the heartbeat shape best fits the signal. The method was validated with overnight recordings from 46 subjects in varying setups (sleep clinic, home, single bed, double bed, two sensor types). The mean beat-to-beat interval error was 13 ms and on an average 54% of the beat-to-beat intervals were detected. The method is part of a home-use e-health system for an unobtrusive sleep measurement.

Sleep and wake classification with actigraphy and respiratory effort using dynamic warping

This paper proposes the use of dynamic warping (DW) methods for improving automatic sleep and wake classification using actigraphy and respiratory effort. DW is an algorithm that finds an optimal nonlinear alignment between two series allowing scaling and shifting. It is widely used to quantify (dis)similarity between two series. To compare the respiratory effort between sleep and wake states by means of (dis)similarity, we constructed two novel features based on DW. For a given epoch of a respiratory effort recording, the features search for the optimally aligned epoch within the same recording in time and frequency domain. This is expected to yield a high (or low) similarity score when this epoch is sleep (or wake). Since the comparison occurs throughout the entire-night recording of a subject, it may reduce the effects of within- and between-subject variations of the respiratory effort, and thus help discriminate between sleep and wake states. The DW-based features were evaluated using a linear discriminant classifier on a dataset of 15 healthy subjects. Results show that the DW-based features can provide a Cohen's Kappa coefficient of agreement κ = 0.59 which is significantly higher than the existing respiratory-based features and is comparable to actigraphy. After combining the actigraphy and the DW-based features, the classifier achieved a κ of 0.66 and an overall accuracy of 95.7%, outperforming an earlier actigraphy- and respiratory-based feature set ( κ = 0.62). The results are also comparable with those obtained using an actigraphy- and cardiorespiratory-based feature set but have the important advantage that they do not require an ECG signal to be recorded.

Unobtrusive online monitoring of sleep at home

We describe an online sleep monitoring service, based on unobtrusive ballistocardiography (BCG) measurement in an ordinary bed. The novelty of the system is that the sleep tracking web application is based on measurements from a fully unobtrusive sensor. The BCG signal is measured with a piezoelectric film sensor under the mattress topper, and sent to the web server for analysis. Heart rate and respiratory variation, activity, sleep stages, and stress reactions are inferred based on the signal. The sleep information is presented to the user along with measurements of the sleeping environment (temperature, noise, luminosity) and user-logged tags (e.g. stress, alcohol, exercise). The approach is designed for long-term use at home, allowing users to follow the development of their sleep over months and years. The service has also a medical use, as sleep disorder patients can be measured for long periods before and after interventions.

Application of recurrence quantification analysis to automatically estimate infant sleep states using a single channel of respiratory data

Previous work has identified that non-linear variables calculated from respiratory data vary between sleep states, and that variables derived from the non-linear analytical tool recurrence quantification analysis (RQA) are accurate infant sleep state discriminators. This study aims to apply these discriminators to automatically classify 30 s epochs of infant sleep as REM, non-REM and wake. Polysomnograms were obtained from 25 healthy infants at 2 weeks, 3, 6 and 12 months of age, and manually sleep staged as wake, REM and non-REM. Inter-breath interval data were extracted from the respiratory inductive plethysmograph, and RQA applied to calculate radius, determinism and laminarity. Time-series statistic and spectral analysis variables were also calculated. A nested cross-validation method was used to identify the optimal feature subset, and to train and evaluate a linear discriminant analysis-based classifier. The RQA features radius and laminarity and were reliably selected. Mean agreement was 79.7, 84.9, 84.0 and 79.2 % at 2 weeks, 3, 6 and 12 months, and the classifier performed better than a comparison classifier not including RQA variables. The performance of this sleep-staging tool compares favourably with inter-human agreement rates, and improves upon previous systems using only respiratory data. Applications include diagnostic screening and population-based sleep research.

Automated determination of wakefulness and sleep in rats based on non-invasively acquired measures of movement and respiratory activity

The current standard for monitoring sleep in rats requires labor intensive surgical procedures and the implantation of chronic electrodes which have the potential to impact behavior and sleep. With the goal of developing a non-invasive method to determine sleep and wakefulness, we constructed a non-contact monitoring system to measure movement and respiratory activity using signals acquired with pulse Doppler radar and from digitized video analysis. A set of 23 frequency and time-domain features were derived from these signals and were calculated in 10s epochs. Based on these features, a classification method for automated scoring of wakefulness, non-rapid eye movement sleep (NREM) and REM in rats was developed using a support vector machine (SVM). We then assessed the utility of the automated scoring system in discriminating wakefulness and sleep by comparing the results to standard scoring of wakefulness and sleep based on concurrently recorded EEG and EMG. Agreement between SVM automated scoring based on selected features and visual scores based on EEG and EMG were approximately 91% for wakefulness, 84% for NREM and 70% for REM. The results indicate that automated scoring based on non-invasively acquired movement and respiratory activity will be useful for studies requiring discrimination of wakefulness and sleep. However, additional information or signals will be needed to improve discrimination of NREM and REM episodes within sleep.

Adaptive sleep-wake discrimination for wearable devices

Sleep/wake classification systems that rely on physiological signals suffer from intersubject differences that make accurate classification with a single, subject-independent model difficult. To overcome the limitations of intersubject variability, we suggest a novel online adaptation technique that updates the sleep/wake classifier in real time. The objective of the present study was to evaluate the performance of a newly developed adaptive classification algorithm that was embedded on a wearable sleep/wake classification system called SleePic. The algorithm processed ECG and respiratory effort signals for the classification task and applied behavioral measurements (obtained from accelerometer and press-button data) for the automatic adaptation task. When trained as a subject-independent classifier algorithm, the SleePic device was only able to correctly classify 74.94 ± 6.76% of the human-rated sleep/wake data. By using the suggested automatic adaptation method, the mean classification accuracy could be significantly improved to 92.98 ± 3.19%. A subject-independent classifier based on activity data only showed a comparable accuracy of 90.44 ± 3.57%. We demonstrated that subject-independent models used for online sleep-wake classification can successfully be adapted to previously unseen subjects without the intervention of human experts or off-line calibration.

Sleep/wake detection based on cardiorespiratory signals and actigraphy

We investigated the potential of adding cardiac and respiratory activity information to actigraphy for sleep-wake staging. A dataset of 35 recordings with full polysomnography and actigraphy was used to assess the performance of an automated sleep/wake Bayesian classifier using electrocardiogram, inductance plethysmogram estimate of respiratory effort and actigraphy. The best performance was achieved with the linear discriminant model that provided an agreement of Cohen's kappa=0.62 for one of the configurations of the classifier, corresponding to an accuracy of 86.8%, a sensitivity of 66.9% and a specificity of 93.1%. It shows that combining different vital signs for a home sleep-wake staging system could be a promising approach.