Sleep Quality Prediction From Wearable Data Using Deep Learning

Predicting long-term sleep deprivation using wearable sensors and health surveys

Sufficient sleep is essential for individual well-being. Inadequate sleep has been shown to have significant negative impacts on our attention, cognition, and mood. The measurement of sleep from in-bed physiological signals has progressed to where commercial devices already incorporate this functionality. However, the prediction of sleep duration from previous awake activity is less studied. Previous studies have used daily exercise summaries, actigraph data, and pedometer data to predict sleep during individual nights. Building upon these, this article demonstrates how to predict a person's long-term average sleep length over the course of 30 days from Fitbit-recorded physical activity data alongside self-report surveys. Recursive Feature Elimination with Random Forest (RFE-RF) is used to extract the feature sets used by the machine learning models, and sex differences in the feature sets and performances of different machine learning models are then examined. The feature selection process demonstrates that previous sleep patterns and physical exercise are the most relevant kind of features for predicting sleep. Personality and depression metrics were also found to be relevant. When attempting to classify individuals as being long-term sleep-deprived, good performance was achieved across both the male, female, and combined data sets, with the highest-performing model achieving an AUC of 0.9762. The best-performing regression model for predicting the average nightly sleep time achieved an R-squared of 0.6861, with other models achieving similar results. When attempting to predict if a person who previously was obtaining sufficient sleep would become sleep-deprived, the best-performing model obtained an AUC of 0.9448.

Hybrid Integration of Wearable Devices for Physiological Monitoring

Wearable devices can provide timely, user-friendly, non- or minimally invasive, and continuous monitoring of human health. Recently, multidisciplinary scientific communities have made significant progress regarding fully integrated wearable devices such as sweat wearable sensors, saliva sensors, and wound sensors. However, the translation of these wearables into markets has been slow due to several reasons associated with the poor system-level performance of integrated wearables. The wearability consideration for wearable devices compromises many properties of the wearables. Besides, the limited power capacity of wearables hinders continuous monitoring for extended duration. Furthermore, peak-power operations for intensive computations can quickly create thermal issues in the compact form factor that interfere with wearability and sensor operations. Moreover, wearable devices are constantly subjected to environmental, mechanical, chemical, and electrical interferences and variables that can invalidate the collected data. This generates the need for sophisticated data analytics to contextually identify, include, and exclude data points per multisensor fusion to enable accurate data interpretation. This review synthesizes the challenges surrounding the wearable device integration from three aspects in terms of hardware, energy, and data, focuses on a discussion about hybrid integration of wearable devices, and seeks to provide comprehensive guidance for designing fully functional and stable wearable devices.

Extracting Stress-Related EEG Patterns From Pre-Sleep EEG for Forecasting Slow-Wave Sleep Deficiency

Sleep is vital to our daily activity. Lack of proper sleep can impair functionality and overall health. While stress is known for its detrimental impact on sleep quality, the precise effect of pre-sleep stress on subsequent sleep structure remains unknown. This study introduced a novel approach to study the pre-sleep stress effect on sleep structure, specifically slow-wave sleep (SWS) deficiency. To achieve this, we selected forehead resting EEG immediately before and upon sleep onset to extract stress-related neurological markers through power spectra and entropy analysis. These markers include beta/delta correlation, alpha asymmetry, fuzzy entropy (FuzzEn) and spectral entropy (SpEn). Fifteen subjects were included in this study. Our results showed that subjects lacking SWS often exhibited signs of stress in EEG, such as an increased beta/delta correlation, higher alpha asymmetry, and increased FuzzEn in frontal EEG. Conversely, individuals with ample SWS displayed a weak beta/delta correlation and reduced FuzzEn. Finally, we employed several supervised learning models and found that the selected neurological markers can predict subsequent SWS deficiency. Our investigation demonstrated that the classifiers could effectively predict varying levels of slow-wave sleep (SWS) from pre-sleep EEG segments, achieving a mean balanced accuracy surpassing 0.75. The SMOTE-Tomek resampling method could improve the performance to 0.77. This study suggests that stress-related neurological markers derived from pre-sleep EEG can effectively predict SWS deficiency. Such information can be integrated with existing sleep-improving techniques to provide a personalized sleep forecasting and improvement solution.

Applications and Challenges of Telemedicine: Privacy-Preservation as a Case Study

Today, technology has an important impact on the development of medical services, especially during the outbreak of COVID-19. Telemedicine, known by terms such as telehealth and digital health, refers to the utilization of technology to provide health care services at a distance that leads to improved monitoring, detecting and treatment of disease, and provision of individual care. It has been considered in various fields such as radiology, cardiology, pulmonology, psychiatry, emergency care and surgery. The most important advantages of using telemedicine are saving time for the doctor and the patient, reducing the cost of multiple visits to the doctor, reducing the spread of contagious diseases and caring for patients who cannot see a doctor, such as the elderly. In this paper, we review the research in the field of applying telemedicine, as well as its advantages and disadvantages. Next, we discuss the challenges in the field of using telemedicine which are privacy preserving, data security, cost of infrastructures, lack of physical examination and responsibility for patients' compensation. One of the most important challenges is privacy preserving of patients' information during transmission and process. We categorize and compare the various methods that have been proposed to protect peoples' privacy.

Can Sleep Quality Attributes be Predicted from Physical Activity in Everyday Settings?

Sleep is crucial for physical, mental, and emotional well-being. Physical activity and sleep are known to be interrelated; however, limited research has been performed to investigate their interactions in long-term. Conventional studies have presented sleep quality prediction, focusing on a single sleep quality aspect, such as sleep efficiency. In addition, the relationship between daily physical activity and sleep quality has yet to be explored, despite physical activities being utilized in previous studies for sleep quality prediction. In this paper, we develop an Extreme Gradient boosting method to predict sleep duration, sleep efficiency, and deep sleep based on users' daily activity information collected from wearable devices. Our model is trained and tested using data collected with an OURA ring from 34 pregnant mothers for six months under free-living conditions. Our finding shows an accuracy of 90.58%, 95.38%, and 91.45% for sleep duration, efficiency, and deep sleep, respectively. Moreover, we assess the contribution of each physical activity parameter to the prediction results using the Shapley Additive Explanations method. Our results indicate that sedentary time is the most influential parameter for sleep duration prediction, while the inactive time feature (e.g., resting or lying down) has a strong negative relationship with sleep efficiency, and the pregnancy week is the most critical parameter for deep sleep prediction.

Enhancing user awareness on inferences obtained from fitness trackers data

In the IoT era, sensitive and non-sensitive data are recorded and transmitted to multiple service providers and IoT platforms, aiming to improve the quality of our lives through the provision of high-quality services. However, in some cases these data may become available to interested third parties, who can analyse them with the intention to derive further knowledge and generate new insights about the users, that they can ultimately use for their own benefit. This predicament raises a crucial issue regarding the privacy of the users and their awareness on how their personal data are shared and potentially used. The immense increase in fitness trackers use has further increased the amount of user data generated, processed and possibly shared or sold to third parties, enabling the extraction of further insights about the users. In this work, we investigate if the analysis and exploitation of the data collected by fitness trackers can lead to the extraction of inferences about the owners routines, health status or other sensitive information. Based on the results, we utilise the PrivacyEnhAction privacy tool, a web application we implemented in a previous work through which the users can analyse data collected from their IoT devices, to educate the users about the possible risks and to enable them to set their user privacy preferences on their fitness trackers accordingly, contributing to the personalisation of the provided services, in respect of their personal data.

Feasibility of wearable devices and machine learning for sleep classification in children with Rett syndrome: A pilot study

Sleep is vital to many processes involved in the well-being and health of children; however, it is estimated that 80% of children with Rett syndrome suffer from sleep disorders. Caregiver reports and questionnaires, which are the current method of studying sleep, are prone to observer bias and missed information. Polysomnography is considered the gold standard for sleep analysis but is labor and cost-intensive and limits the frequency of data collection for sleep disorder studies. Wearable digital health technologies, such as actigraphy devices, have shown potential and feasibility as a method for sleep analysis in Rett syndrome, but have not been validated against polysomnography. Furthermore, the collected accelerometer data has limitations due to the rigidity, periodic limb movement, and involuntary muscle contractions prevalent in Rett syndrome. Heart rate and electrodermal activity, along with other physiological signals, have been linked to sleep stages and can be utilized with machine learning to provide better resistance to noise and false positives than actigraphy. This research aims to address the gap in Rett syndrome sleep analysis by comparing the performance of a machine learning model utilizing both accelerometer data and physiological data features to the gold-standard polysomnography for sleep analysis in Rett syndrome. Our analytical validation pilot study (n = 7) found that using physiological and accelerometer features, our machine learning models can differentiate between awake, non-rapid eye movement sleep, and rapid eye movement sleep in Rett syndrome children with an accuracy of 85.1% when using an individual model. Additionally, this work demonstrates that it is feasible to use digital health technologies in Rett syndrome, even at a young age, without data loss or interference from repetitive movements that are characteristic of Rett syndrome.

Evaluating machine learning models to classify occupants' perceptions of their indoor environment and sleep quality from indoor air quality

A variety of factors can affect a person's perception of their environment and health, but one factor that is often overlooked in indoor settings is the air quality. To address this gap, we develop and evaluate four Machine Learning (ML) models on two disparate datasets using Indoor Air Quality (IAQ) parameters as primary features and components of self-reported IAQ satisfaction and sleep quality as target variables. In each case, we compare models to each other as well as to a simple model that always predicts the majority outcome. In the first analysis, we use open-source data collected from 93 California residences to predict occupant's satisfaction with their indoor environment. Results indicate building ventilation rate, Relative Humidity (RH), and formaldehyde are most influential when predicting IAQ perception and do so with an accuracy greater than the simplified model. The second analysis uses IAQ data gathered from a field study we conducted with 20 participants over 11 weeks to train similar models. We obtain accuracy and F1 scores similar to the simplified model where PM_2.5 and TVOCs represent the most important predictors. Our results underscore the ability of IAQ to affect a person's perception of their built environment and health and highlight the utility of ML models to explore the strength of these relationships.Implications: The results from this study show that two outcome variables - occupant's indoor air quality (IAQ) satisfaction and perceived sleep quality - are related to the measured IAQ parameters but not heavily influenced by typical values measured in apartments and homes. This study highlights the ability of machine learning models as exploratory analysis tools to determine underlying relationships within and across datasets in addition to understanding the importance of certain features on the outcome variable. We compare four different models and find that the random forest classifier has the best performance in both analysis on IAQ satisfaction and perceived sleep quality. It is a suitable model for predicting IAQ-related subjective metrics and also provides value insight into the feature importance of the IAQ parameters. The accuracy of any of these machine learning models in predicting occupants' comfort or sleep quality is limited by the dataset size, how data is collected, and range of data. This study identifies the factors that are important to IAQ perception: ventilation rate, relative humidity, and concentrations of formaldehyde, NO2, and particulate matter. It indicates that sensors that can measure these variables are necessary for future, related studies that model occupants' IAQ satisfaction. However, this study does not find strong relationships between any of the IAQ parameters measured in this study and perceived sleep quality despite the logical pathway between these many pollutants and respiratory issues. A prediction model of IAQ perception or sleep quality can be integrated into home management systems to automatically adjust building operations such as ventilation rates in smart buildings. Once buildings are equipped with a network of low-cost sensors that measure concentrations of pollutants and operating conditions of the ventilation system, the prediction model can be used to predict the occupants' comfort and facilitate the control of the ventilation system.

Identifying Sleep Disorders From Search Engine Activity: Combining User-Generated Data With a Clinically Validated Questionnaire

Background

Sleep disorders are experienced by up to 40% of the population but their diagnosis is often delayed by the availability of specialists.

Objective

We propose the use of search engine activity in conjunction with a validated web-based sleep questionnaire to facilitate wide-scale screening of prevalent sleep disorders.

Methods

Search advertisements offering a web-based sleep disorder screening questionnaire were shown on the Bing search engine to individuals who indicated an interest in sleep disorders. People who clicked on the advertisements and completed the sleep questionnaire were identified as being at risk for 1 of 4 common sleep disorders. A machine learning algorithm was applied to previous search engine queries to predict their suspected sleep disorder, as identified by the questionnaire.

Results

A total of 397 users consented to participate in the study and completed the questionnaire. Of them, 132 had sufficient past query data for analysis. Our findings show that diurnal patterns of people with sleep disorders were shifted by 2-3 hours compared to those of the controls. Past query activity was predictive of sleep disorders, approaching an area under the receiver operating characteristic curve of 0.62-0.69, depending on the sleep disorder.

Conclusions

Targeted advertisements can be used as an initial screening tool for people with sleep disorders. However, search engine data are seemingly insufficient as a sole method for screening. Nevertheless, we believe that evaluable web-based information, easily collected and processed with little effort on part of the physician and with low burden on the individual, can assist in the diagnostic process and possibly drive people to seek sleep assessment and diagnosis earlier than they currently do.

Associations Between Sleep and Metabolic Outcomes in Preadolescent Children

Context

Growing evidence suggests an important role for sleep for the metabolic health of children.

Objective

We aimed to determine how sleep is related to insulin sensitivity, insulin secretion, beta-cell function, and adiposity (BMI z-scores, body fat %, waist to height ratio) using objectively measured sleep and oral glucose tolerance test (OGTT)-derived measures.

Methods

Sixty-two children aged 7-11 years, born at Kaiser Permanente Southern California, wore wrist accelerometers for 7 days to objectively measure sleep, completed an OGTT, and had anthropometric measures (height [cm], weight [kg], waist [cm], body fat [%]) collected. Using linear regression, associations between Matsuda insulin sensitivity index (ISI), insulinogenic index (IGI), disposition index (DI), BMI z-score, waist to height ratio, and body fat % with sleep parameters [total sleep time (TST; min), sleep efficiency (SE; %), time in bed (TIB; min), wake after sleep onset (WASO; min), and sleep latency (SL; min)] were assessed. Body fat % was tested as a mediator of the relationship between TST and ISI.

Results

Longer TST was associated with better insulin sensitivity (P = 0.02), but not after adjusting for body fat %. Sleep parameters were not associated with IGI or DI. Longer TST was associated with lower % body fat (P = 0.01) and lower waist-to-height-ratios (P = 0.05). Body fat % explained 62% (P = 0.01) of the relationship between TST and ISI. Longer TIB was associated with lower adiposity measures (P < 0.05). There were no associations between SE, WASO, or SL and metabolic outcomes.

Conclusion

Objectively measured sleep duration was associated with lower adiposity, and the relationship between sleep duration and ISI appeared partly through adiposity levels in preadolescent children. Longer sleep duration may be important for metabolic health.

Smart Consumer Wearables as Digital Diagnostic Tools: A Review

The increasing usage of smart wearable devices has made an impact not only on the lifestyle of the users, but also on biological research and personalized healthcare services. These devices, which carry different types of sensors, have emerged as personalized digital diagnostic tools. Data from such devices have enabled the prediction and detection of various physiological as well as psychological conditions and diseases. In this review, we have focused on the diagnostic applications of wrist-worn wearables to detect multiple diseases such as cardiovascular diseases, neurological disorders, fatty liver diseases, and metabolic disorders, including diabetes, sleep quality, and psychological illnesses. The fruitful usage of wearables requires fast and insightful data analysis, which is feasible through machine learning. In this review, we have also discussed various machine-learning applications and outcomes for wearable data analyses. Finally, we have discussed the current challenges with wearable usage and data, and the future perspectives of wearable devices as diagnostic tools for research and personalized healthcare domains.

Modeling Sleep Quality Depending on Objective Actigraphic Indicators Based on Machine Learning Methods

According to data from the World Health Organization and medical research centers, the frequency and severity of various sleep disorders, including insomnia, are increasing steadily. This dynamic is associated with increased daily stress, anxiety, and depressive disorders. Poor sleep quality affects people's productivity and activity and their perception of quality of life in general. Therefore, predicting and classifying sleep quality is vital to improving the quality and duration of human life. This study offers a model for assessing sleep quality based on the indications of an actigraph, which was used by 22 participants in the experiment for 24 h. Objective indicators of the actigraph include the amount of time spent in bed, sleep duration, number of awakenings, and duration of awakenings. The resulting classification model was evaluated using several machine learning methods and showed a satisfactory accuracy of approximately 80-86%. The results of this study can be used to treat sleep disorders, develop and design new systems to assess and track sleep quality, and improve existing electronic devices and sensors.

A Participatory Design Approach to Develop Visualization of Wearable Actigraphy Data for Health Care Professionals: Case Study in Qatar

Background

Several tools have been developed for health care professionals to monitor the physical activity of their patients, but most of these tools have been considering only the needs of users in North American and European countries and applicable for only specific analytic tasks. To our knowledge, no research study has utilized the participatory design (PD) approach in the Middle East region to develop such tools, involving all the stakeholders in the product development phases, and no clear use cases have been derived from such studies that could serve future development in the field.

Objective

This study aims to develop an interactive visualization tool (ActiVis) to support local health care professionals in monitoring the physical activity of their patients measured through wearable sensors, with the overall objective of improving the health of the Qatari population.

Methods

We used PD and user-centered design methodologies to develop ActiVis, including persona development, brainwriting, and heuristic walkthrough as part of user evaluation workshops; and use cases, heuristic walkthrough, interface walkthrough, and survey as part of expert evaluation sessions.

Results

We derived and validated 6 data analysis use cases targeted at specific health care professionals from a collaborative design workshop and an expert user study. These use cases led to improving the design of the ActiVis tool to support the monitoring of patients’ physical activity by nurses and family doctors. The ActiVis research prototype (RP) compared favorably with the Fitbit Dashboard, showing the importance of design tools specific to end users’ needs rather than relying on repurposing existing tools designed for other types of users. The use cases we derived happen to be culturally agnostic, despite our assumption that the local Muslim and Arabic culture could impact the design of such visualization tools. At last, taking a step back, we reflect on running collaborative design sessions in a multicultural environment and oil-based economy.

Conclusions

Beyond the development of the ActiVis tool, this study can serve other visualization and human–computer interaction designers in the region to prepare their design projects and encourage health care professionals to engage with designers and engineers to improve the tools they use for supporting their daily routine. The development of the ActiVis tool for nurses, and other visualization tools specific to family doctors and clinician researchers, is still ongoing and we plan to integrate them into an operational platform for health care professionals in Qatar in the near future.

Importance of Getting Enough Sleep and Daily Activity Data to Assess Variability: Longitudinal Observational Study

Background

The gold standard measurement for recording sleep is polysomnography performed in a hospital environment for 1 night. This requires individuals to sleep with a device and several sensors attached to their face, scalp, and body, which is both cumbersome and expensive. Self-trackers, such as wearable sensors (eg, smartwatch) and nearable sensors (eg, sleep mattress), can measure a broad range of physiological parameters related to free-living sleep conditions; however, the optimal duration of such a self-tracker measurement is not known. For such free-living sleep studies with actigraphy, 3 to 14 days of data collection are typically used.

Objective

The primary goal of this study is to investigate if 3 to 14 days of sleep data collection is sufficient while using self-trackers. The secondary goal is to investigate whether there is a relationship among sleep quality, physical activity, and heart rate. Specifically, we study whether individuals who exhibit similar activity can be clustered together and to what extent the sleep patterns of individuals in relation to seasonality vary.

Methods

Data on sleep, physical activity, and heart rate were collected over 6 months from 54 individuals aged 52 to 86 years. The Withings Aura sleep mattress (nearable; Withings Inc) and Withings Steel HR smartwatch (wearable; Withings Inc) were used. At the individual level, we investigated the consistency of various physical activities and sleep metrics over different time spans to illustrate how sensor data from self-trackers can be used to illuminate trends. We used exploratory data analysis and unsupervised machine learning at both the cohort and individual levels.

Results

Significant variability in standard metrics of sleep quality was found between different periods throughout the study. We showed specifically that to obtain more robust individual assessments of sleep and physical activity patterns through self-trackers, an evaluation period of >3 to 14 days is necessary. In addition, we found seasonal patterns in sleep data related to the changing of the clock for daylight saving time.

Conclusions

We demonstrate that >2 months’ worth of self-tracking data are needed to provide a representative summary of daily activity and sleep patterns. By doing so, we challenge the current standard of 3 to 14 days for sleep quality assessment and call for the rethinking of standards when collecting data for research purposes. Seasonal patterns and daylight saving time clock change are also important aspects that need to be taken into consideration when choosing a period for collecting data and designing studies on sleep. Furthermore, we suggest using self-trackers (wearable and nearable ones) to support longer-term evaluations of sleep and physical activity for research purposes and, possibly, clinical purposes in the future.

An app to classify a 5-year survival in patients with breast cancer using the convolutional neural networks (CNN) in Microsoft Excel: Development and usability study

BACKGROUND

Breast cancer (BC) is the most common malignant cancer in women. A predictive model is required to predict the 5-year survival in patients with BC (5YSPBC) and improve the treatment quality by increasing their survival rate. However, no reports in literature about apps developed and designed in medical practice to classify the 5YSPBC. This study aimed to build a model to develop an app for an automatically accurate classification of the 5YSPBC.

METHODS

A total of 1810 patients with BC were recruited in a hospital in Taiwan from the secondary data with codes on 53 characteristic variables that were endorsed by professional staff clerks as of December 31, 2019. Five models (i.e., revolution neural network [CNN], artificial neural network, Naïve Bayes, K-nearest Neighbors Algorithm, and Logistic regression) and 3 tasks (i.e., extraction of feature variables, model comparison in accuracy [ACC] and stability, and app development) were performed to achieve the goal of developing an app to predict the 5YSPBC. The sensitivity, specificity, and receiver operating characteristic curve (area under ROC curve) on models across 2 scenarios of training (70%) and testing (30%) sets were compared. An app predicting the 5YSPBC was developed involving the model estimated parameters for a website assessment.

RESULTS

We observed that the 15-variable CNN model yields higher ACC rates (0.87 and 0.86) with area under ROC curves of 0.80 and 0.78 (95% confidence interval 0.78-82 and 0.74-81) based on 1357 training and 540 testing cases an available app for patients predicting the 5YSPBC was successfully developed and demonstrated in this study.

CONCLUSION

The 15-variable CNN model with 38 parameters estimated using CNN for improving the ACC of the 5YSPBC has been particularly demonstrated in Microsoft Excel. An app developed for helping clinicians assess the 5YSPBC in clinical settings is required for application in the future.

Prediction of good sleep with physical activity and light exposure: a preliminary study

STUDY OBJECTIVES

Cognitive behavioral treatment for insomnia is performed under the premise that feedback provided by evaluation of sleep diaries written by patients will result in good sleep. The sleep diary is essential for behavior therapy and sleep hygiene education. However, limitations include subjectivity and laborious input. We aimed to develop an artificial intelligence sleep prediction model and to find factors associated with good sleep using a wrist-worn actigraphy device.

METHODS

We enrolled 109 participants who reported having no sleep disturbances. We developed a sleep prediction model using 733 days of actigraphy data of physical activity and light exposure. Twenty-four sleep prediction models were developed based on different data sources (actigraphy alone, sleep diary alone, or combined data), different durations of data (1 or 2 days), and different analysis methods (extreme gradient boosting, convolutional neural network, long short-term memory, logistic regression analysis). The outcome measure of "good sleep" was defined as ≥90% sleep efficiency.

RESULTS

Actigraphy model performance was comparable to sleep diary model performance. Two-day models generally performed better than 1-day models. Among all models, the 2-day, combined (actigraphy and sleep diary), extreme gradient boosting model had the best performance for predicting good sleep (accuracy=0.69, area under the curve=0.70).

CONCLUSIONS

The findings suggested that it is possible to develop automated sleep models with good predictive performance. Further research including patients with insomnia is needed for clinical application.

Advancing understanding of dietary and movement behaviours in an Asian population through real-time monitoring: Protocol of the Continuous Observations of Behavioural Risk Factors in Asia study (COBRA)

Background

Modifiable risk factors for non-communicable diseases, including eating an unhealthy diet and being physically inactive, are influenced by complex and dynamic interactions between people and their social and physical environment. Therefore, understanding patterns and determinants of these risk factors as they occur in real life is essential to enable the design of precision public health interventions.

Objective

This paper describes the protocol for the Continuous Observations of Behavioural Risk Factors in Asia study (COBRA). The study uses real-time data capture methods to gain a comprehensive understanding of eating and movement behaviours, including how these differ by socio-demographic characteristics and are shaped by people's interaction with their social and physical environment.

Methods

COBRA is an observational study in free-living conditions. We will recruit 1500 adults aged 21-69 years from a large prospective cohort study. Real-time data capture methods will be used for nine consecutive days: an ecological momentary assessment app with a global positioning system enabled to collect location data, accelerometers to measure movement, and wearable sensors to monitor blood glucose levels. Participants receive six EMA surveys per day between 8 a.m. and 9.30 p.m. to capture information on behavioural risk factors including eating behaviours and diet composition movement behaviours (physical activity, sedentary behaviour, sleep), and related contextual factors. The second wave of ecological momentary assessment surveys with a global positioning system enabled will be sent 6 months later. Data will be analysed using generalised linear models to examine associations between behavioural risk factors and contextual determinants.

Discussion

Findings from this study will advance our understanding of dietary and movement behaviours as they occur in real-life and inform the development of personalised interventions to prevent chronic diseases.

An application for classifying perceptions on my health bank in Taiwan using convolutional neural networks and web-based computerized adaptive testing: A development and usability study

BACKGROUND

The classification of a respondent's opinions online into positive and negative classes using a minimal number of questions is gradually changing and helps turn techniques into practices. A survey incorporating convolutional neural networks (CNNs) into web-based computerized adaptive testing (CAT) was used to collect perceptions on My Health Bank (MHB) from users in Taiwan. This study designed an online module to accurately and efficiently turn a respondent's perceptions into positive and negative classes using CNNs and web-based CAT.

METHODS

In all, 640 patients, family members, and caregivers with ages ranging from 20 to 70 years who were registered MHB users were invited to complete a 3-domain, 26-item, 5-category questionnaire asking about their perceptions on MHB (PMHB26) in 2019. The CNN algorithm and k-means clustering were used for dividing respondents into 2 classes of unsatisfied and satisfied classes and building a PMHB26 predictive model to estimate parameters. Exploratory factor analysis, the Rasch model, and descriptive statistics were used to examine the demographic characteristics and PMHB26 factors that were suitable for use in CNNs and Rasch multidimensional CAT (MCAT). An application was then designed to classify MHB perceptions.

RESULTS

We found that 3 construct factors were extracted from PMHB26. The reliability of PMHB26 for each subscale beyond 0.94 was evident based on internal consistency and stability in the data. We further found the following: the accuracy of PMHB26 with CNN yields a higher accuracy rate (0.98) with an area under the curve of 0.98 (95% confidence interval, 0.97-0.99) based on the 391 returned questionnaires; and for the efficiency, approximately one-third of the items were not necessary to answer in reducing the respondents' burdens using Rasch MCAT.

CONCLUSIONS

The PMHB26 CNN model, combined with the Rasch online MCAT, is recommended for improving the accuracy and efficiency of classifying patients' perceptions of MHB utility. An application developed for helping respondents self-assess the MHB cocreation of value can be applied to other surveys in the future.

Correlates of poor sleep based upon wrist actigraphy data in bipolar disorder

BACKGROUND

Wrist-worn actigraphy can objectively measure sleep, and has advantages over self-report, particularly for people with Bipolar Disorder (BD) for whom self-reports might be influenced by affect. Clinically useful data reduction approaches are needed to explore these complex data.

METHODS

We created a composite score of sleep metrics in BD based on 51 BD and 80 healthy comparison (HC) participants. Subjects wore an actigraph for up to 14 consecutive 24-h periods, and we assessed total sleep time (TST), wake after sleep onset (WASO), percent sleep (PS), and number of awakenings (NA). We focused on participants who had at least 5 nights of actigraphy data. We computed z-scores for within-person means of sleep measures for BD subjects versus HCs, which were averaged to create a composite measure. We correlated this composite with participant characteristics, and used LASSO regression to identify sleep measures best explaining variability in identified correlates.

RESULTS

Sleep measures and the composite did not differ between BDs and HCs; however, there was considerable variability in z-scores among those with BD. In BDs, the composite score was higher in women (t₍₄₉₎ = 2.28, p = 0.027) and those who were employed (t₍₃₄₎ = 2.34, p = 0.025), and positively correlated with medication load (r = 0.41, p = 0.003) while negatively correlated with Young Mania Rating Scale (YMRS; r = -0.35, p = 0.030). In LASSO regression, TST and NA best explained medication load while PS best explained employment and YMRS.

CONCLUSION

While a composite score of sleep metrics may provide useful information about sleep quality globally, our findings suggest that selection of theory-driven sleep measures may be more clinically meaningful.

Smart Healthcare System Based on Cloud-Internet of Things and Deep Learning

Introduction

Health monitoring and remote diagnosis can be realized through Smart Healthcare. In view of the existing problems such as simple measurement parameters of wearable devices, huge computing pressure of cloud servers, and lack of individualization of diagnosis, a novel Cloud-Internet of Things (C-IOT) framework for medical monitoring is put forward.

Methods

Smart phones are adopted as gateway devices to achieve data standardization and preprocess to generate health gray-scale map uploaded to the cloud server. The cloud server realizes the business logic processing and uses the deep learning model to carry out the gray-scale map calculation of health parameters. A deep learning model based on the convolution neural network (CNN) is constructed, in which six volunteers are selected to participate in the experiment, and their health data are marked by private doctors to generate initial data set.

Results

Experimental results show the feasibility of the proposed framework. The test data set is used to test the CNN model after training; the forecast accuracy is over 77.6%.

Conclusion

The CNN model performs well in the recognition of health status. Collectively, this Smart Healthcare System is expected to assist doctors by improving the diagnosis of health status in clinical practice.

The Digital Neurologic Examination

Digital health has been rapidly thrust into the forefront of care delivery. Poised to extend the clinician's reach, a new set of examination tools will redefine neurologic and neurosurgical care, serving as the basis for the digital neurologic examination. We describe its components and review specific technologies, which move beyond traditional video-based telemedicine encounters and include separate digital tools. A future suite of these clinical assessment technologies will blur the lines between history taking, examination, and remote monitoring. Prior to full-scale implementation, however, much more investigation is needed. Because of the nascent state of the technologies, researchers, clinicians, and developers should establish digital neurologic examination requirements in order to maximize its impact.

Predicting Active NBA Players Most Likely to Be Inducted into the Basketball Hall of Famers Using Artificial Neural Networks in Microsoft Excel: Development and Usability Study

The prediction of whether active NBA players can be inducted into the Hall of Fame (HOF) is interesting and important. However, no such research have been published in the literature, particularly using the artificial neural network (ANN) technique. The aim of this study is to build an ANN model with an app for automatic prediction and classification of HOF for NBA players. We downloaded 4728 NBA players’ data of career stats and accolades from the website at basketball-reference.com. The training sample was collected from 85 HOF members and 113 retired Non-HOF players based on completed data and a longer career length (≥15 years). Featured variables were taken from the higher correlation coefficients (<0.1) with HOF and significant deviations apart from the two HOF/Non-HOF groups using logistical regression. Two models (i.e., ANN and convolutional neural network, CNN) were compared in model accuracy (e.g., sensitivity, specificity, area under the receiver operating characteristic curve, AUC). An app predicting HOF was then developed involving the model’s parameters. We observed that (1) 20 feature variables in the ANN model yielded a higher AUC of 0.93 (95% CI 0.93–0.97) based on the 198-case training sample, (2) the ANN performed better than CNN on the accuracy of AUC (= 0.91, 95% CI 0.87–0.95), and (3) an ready and available app for predicting HOF was successfully developed. The 20-variable ANN model with the 53 parameters estimated by the ANN for improving the accuracy of HOF has been developed. The app can help NBA fans to predict their players likely to be inducted into the HOF and is not just limited to the active NBA players.

Nighttime features derived from topic models for classification of patients with COPD

Nighttime symptoms are important indicators of impairment for many diseases and particularly for respiratory diseases such as chronic obstructive pulmonary disease (COPD). The use of wearable sensors to assess sleep in COPD has mainly been limited to the monitoring of limb motions or the duration and continuity of sleep. In this paper we present an approach to concisely describe sleep patterns in subjects with and without COPD. The methodology converts multimodal sleep data into a text representation and uses topic modeling to identify patterns across the dataset composed of more than 6000 assessed nights. This approach enables the discovery of higher level features resembling unique sleep characteristics that are then used to discriminate between healthy subjects and those with COPD and to evaluate patients' disease severity and dyspnea level. Compared to standard features, the discovered latent structures in nighttime data seem to capture important aspects of subjects sleeping behavior related to the effects of COPD and dyspnea.

Machine and Deep Learning in Molecular and Genetic Aspects of Sleep Research

Epidemiological sleep research strives to identify the interactions and causal mechanisms by which sleep affects human health, and to design intervention strategies for improving sleep throughout the lifespan. These goals can be advanced by further focusing on the environmental and genetic etiology of sleep disorders, and by development of risk stratification algorithms, to identify people who are at risk or are affected by, sleep disorders. These studies rely on comprehensive sleep-related data which often contains complex multi-dimensional physiological and molecular measurements across multiple timepoints. Thus, sleep research is well-suited for the application of computational approaches that can handle high-dimensional data. Here, we survey recent advances in machine and deep learning together with the availability of large human cohort studies with sleep data that can jointly drive the next breakthroughs in the sleep-research field. We describe sleep-related data types and datasets, and present some of the tasks in the field that can be targets for algorithmic approaches, as well as the challenges and opportunities in pursuing them.

Deep learning for small and big data in psychiatry

Psychiatry today must gain a better understanding of the common and distinct pathophysiological mechanisms underlying psychiatric disorders in order to deliver more effective, person-tailored treatments. To this end, it appears that the analysis of 'small' experimental samples using conventional statistical approaches has largely failed to capture the heterogeneity underlying psychiatric phenotypes. Modern algorithms and approaches from machine learning, particularly deep learning, provide new hope to address these issues given their outstanding prediction performance in other disciplines. The strength of deep learning algorithms is that they can implement very complicated, and in principle arbitrary predictor-response mappings efficiently. This power comes at a cost, the need for large training (and test) samples to infer the (sometimes over millions of) model parameters. This appears to be at odds with the as yet rather 'small' samples available in psychiatric human research to date (n < 10,000), and the ambition of predicting treatment at the single subject level (n = 1). Here, we aim at giving a comprehensive overview on how we can yet use such models for prediction in psychiatry. We review how machine learning approaches compare to more traditional statistical hypothesis-driven approaches, how their complexity relates to the need of large sample sizes, and what we can do to optimally use these powerful techniques in psychiatric neuroscience.

Human Activity Recognition using Inertial, Physiological and Environmental Sensors: A Comprehensive Survey

In the last decade, Human Activity Recognition (HAR) has become a vibrant research area, especially due to the spread of electronic devices such as smartphones, smartwatches and video cameras present in our daily lives. In addition, the advance of deep learning and other machine learning algorithms has allowed researchers to use HAR in various domains including sports, health and well-being applications. For example, HAR is considered as one of the most promising assistive technology tools to support elderly's daily life by monitoring their cognitive and physical function through daily activities. This survey focuses on critical role of machine learning in developing HAR applications based on inertial sensors in conjunction with physiological and environmental sensors.

Construction and Application of a Medical-Grade Wireless Monitoring System for Physiological Signals at General Wards

Physiological signals can contain abundant personalized information and indicate health status and disease deterioration. However, in current medical practice, clinicians working in the general wards are usually lack of plentiful means and tools to continuously monitor the physiological signals of the inpatients. To address this problem, we here presented a medical-grade wireless monitoring system based on wearable and artificial intelligence technology. The system consists of a multi-sensor wearable device, database servers and user interfaces. It can monitor physiological signals such as electrocardiography and respiration and transmit data wirelessly. We highly integrated the system with the existing hospital information system and explored a set of processes of physiological signal acquisition, storage, analysis, and combination with electronic health records. Multi-scale information extracted from physiological signals and related to the deterioration or abnormality of patients could be shown on the user interfaces, while a variety of reports could be provided daily based on time-series signal processing technology and machine learning to make more information accessible to clinicians. Apart from an initial attempt to implement the system in a realistic clinical environment, we also conducted a preliminary validation of the core processes in the workflow. The heart rate veracity validation of 22 patient volunteers showed that the system had a great consistency with ECG Holter, and bias for heart rate was 0.04 (95% confidence interval: -7.34 to 7.42) beats per minute. The Bland-Altman analysis showed that 98.52% of the points were located between Mean ± 1.96SD. This system has been deployed in the general wards of the Hyperbaric Oxygen Department and Respiratory Medicine Department and has collected more than 1000 cases from the clinic. The whole system will continue to be updated based on clinical feedback. It has been demonstrated that this system can provide reliable physiological monitoring for patients in general wards and has the potential to generate more personalized pathophysiological information related to disease diagnosis and treatment from the continuously monitored physiological data.

Deep-Learning-Based Models for Pain Recognition: A Systematic Review

Traditional standards employed for pain assessment have many limitations. One such limitation is reliability linked to inter-observer variability. Therefore, there have been many approaches to automate the task of pain recognition. Recently, deep-learning methods have appeared to solve many challenges such as feature selection and cases with a small number of data sets. This study provides a systematic review of pain-recognition systems that are based on deep-learning models for the last two years. Furthermore, it presents the major deep-learning methods used in the review papers. Finally, it provides a discussion of the challenges and open issues.

Wearable Device-Independent Next Day Activity and Next Night Sleep Prediction for Rehabilitation Populations

Wearable sensor-based devices are increasingly applied in free-living and clinical settings to collect fine-grained, objective data about activity and sleep behavior. The manufacturers of these devices provide proprietary software that labels the sensor data at specified time intervals with activity and sleep information. If the device wearer has a health condition affecting their movement, such as a stroke, these labels and their values can vary greatly from manufacturer to manufacturer. Consequently, generating outcome predictions based on data collected from patients attending inpatient rehabilitation wearing different sensor devices can be challenging, which hampers usefulness of these data for patient care decisions. In this article, we present a data-driven approach to combining datasets collected from different device manufacturers. With the ability to combine datasets, we merge data from three different device manufacturers to form a larger dataset of time series data collected from 44 patients receiving inpatient therapy services. To gain insights into the recovery process, we use this dataset to build models that predict a patient's next day physical activity duration and next night sleep duration. Using our data-driven approach and the combined dataset, we obtained a normalized root mean square error prediction of 9.11% for daytime physical activity and 11.18% for nighttime sleep duration. Our sleep result is comparable to the accuracy we achieved using the manufacturer's sleep labels (12.26%). Our device-independent predictions are suitable for both point-of-care and remote monitoring applications to provide information to clinicians for customizing therapy services and potentially decreasing recovery time.

An App for Classifying Personal Mental Illness at Workplace Using Fit Statistics and Convolutional Neural Networks: Survey-Based Quantitative Study

BACKGROUND

Mental illness (MI) is common among those who work in health care settings. Whether MI is related to employees' mental status at work is yet to be determined. An MI app is developed and proposed to help employees assess their mental status in the hope of detecting MI at an earlier stage.

OBJECTIVE

This study aims to build a model using convolutional neural networks (CNNs) and fit statistics based on 2 aspects of measures and outfit mean square errors for the automatic detection and classification of personal MI at the workplace using the emotional labor and mental health (ELMH) questionnaire, so as to equip the staff in assessing and understanding their own mental status with an app on their mobile device.

METHODS

We recruited 352 respiratory therapists (RTs) working in Taiwan medical centers and regional hospitals to fill out the 44-item ELMH questionnaire in March 2019. The exploratory factor analysis (EFA), Rasch analysis, and CNN were used as unsupervised and supervised learnings for (1) dividing RTs into 4 classes (ie, MI, false MI, health, and false health) and (2) building an ELMH predictive model to estimate 108 parameters of the CNN model. We calculated the prediction accuracy rate and created an app for classifying MI for RTs at the workplace as a web-based assessment.

RESULTS

We observed that (1) 8 domains in ELMH were retained by EFA, (2) 4 types of mental health (n=6, 63, 265, and 18 located in 4 quadrants) were classified using the Rasch analysis, (3) the 44-item model yields a higher accuracy rate (0.92), and (4) an MI app available for RTs predicting MI was successfully developed and demonstrated in this study.

CONCLUSIONS

The 44-item model with 108 parameters was estimated by using CNN to improve the accuracy of mental health for RTs. An MI app developed to help RTs self-detect work-related MI at an early stage should be made more available and viable in the future.

Deep Learning Identifies Digital Biomarkers for Self-Reported Parkinson's Disease

Large-scale population screening and in-home monitoring for patients with Parkinson's disease (PD) has so far been mainly carried out by traditional healthcare methods and systems. Development of mobile health may provide an independent, future method to detect PD. Current PD detection algorithms will benefit from better generalizability with data collected in real-world situations. In this paper, we report the top-performing smartphone-based method in the recent DREAM Parkinson's Disease Digital Biomarker Challenge for digital diagnosis of PD. Utilizing real-world accelerometer records, this approach differentiated PD from control subjects with an area under the receiver-operating characteristic curve of 0.87 by 3D augmentation of accelerometer records, a significant improvement over other state-of-the-art methods. This study paves the way for future at-home screening of PD and other neurodegenerative conditions affecting movement.

An App for Detecting Bullying of Nurses Using Convolutional Neural Networks and Web-Based Computerized Adaptive Testing: Development and Usability Study

Background

Workplace bullying has been measured in many studies to investigate its effects on mental health issues. However, none have used web-based computerized adaptive testing (CAT) with bully classifications and convolutional neural networks (CNN) for reporting the extent of individual bullying in the workplace.

Objective

This study aims to build a model using CNN to develop an app for automatic detection and classification of nurse bullying-levels, incorporated with online Rasch computerized adaptive testing, to help assess nurse bullying at an earlier stage.

Methods

We recruited 960 nurses working in a Taiwan Ch-Mei hospital group to fill out the 22-item Negative Acts Questionnaire-Revised (NAQ-R) in August 2012. The k-mean and the CNN were used as unsupervised and supervised learnings, respectively, for: (1) dividing nurses into three classes (n=918, 29, and 13 with suspicious mild, moderate, and severe extent of being bullied, respectively); and (2) building a bully prediction model to estimate 69 different parameters. Finally, data were separated into training and testing sets in a proportion of 70:30, where the former was used to predict the latter. We calculated the sensitivity, specificity, and receiver operating characteristic curve (area under the curve [AUC]), along with the accuracy across studies for comparison. An app predicting the respondent bullying-level was developed, involving the model’s 69 estimated parameters and the online Rasch CAT module as a website assessment.

Results

We observed that: (1) the 22-item model yields higher accuracy rates for three categories, with an accuracy of 94% for the total 960 cases, and accuracies of 99% (AUC 0.99; 95% CI 0.99-1.00) and 83% (AUC 0.94; 95% CI 0.82-0.99) for the lower and upper groups (cutoff points at 49 and 66 points) based on the 947 cases and 42 cases, respectively; and (2) the 700-case training set, with 95% accuracy, predicts the 260-case testing set reaching an accuracy of 97. Thus, a NAQ-R app for nurses that predicts bullying-level was successfully developed and demonstrated in this study.

Conclusions

The 22-item CNN model, combined with the Rasch online CAT, is recommended for improving the accuracy of the nurse NAQ-R assessment. An app developed for helping nurses self-assess workplace bullying at an early stage is required for application in the future.

An App Developed for Detecting Nurse Burnouts Using the Convolutional Neural Networks in Microsoft Excel: Population-Based Questionnaire Study

Background

Burnout (BO), a critical syndrome particularly for nurses in health care settings, substantially affects their physical and psychological status, the institute’s well-being, and indirectly, patient outcomes. However, objectively classifying BO levels has not been defined and noticed in the literature.

Objective

The aim of this study is to build a model using the convolutional neural network (CNN) to develop an app for automatic detection and classification of nurse BO using the Maslach Burnout Inventory–Human Services Survey (MBI-HSS) to help assess nurse BO at an earlier stage.

Methods

We recruited 1002 nurses working in a medical center in Taiwan to complete the Chinese version of the 20-item MBI-HSS in August 2016. The k-mean and CNN were used as unsupervised and supervised learnings for dividing nurses into two classes (n=531 and n=471 of suspicious BO+ and BO−, respectively) and building a BO predictive model to estimate 38 parameters. Data were separated into training and testing sets in a proportion 70%:30%, and the former was used to predict the latter. We calculated the sensitivity, specificity, and receiver operating characteristic curve (area under the curve) across studies for comparison. An app predicting respondent BO was developed involving the model’s 38 estimated parameters for a website assessment.

Results

We observed that (1) the 20-item model yields a higher accuracy rate (0.95) with an area under the curve of 0.97 (95% CI 0.94-0.95) based on the 1002 cases, (2) the scheme named matching personal response to adapt for the correct classification in model drives the prior model’s predictive accuracy at 100%, (3) the 700-case training set with 0.96 accuracy predicts the 302-case testing set reaching an accuracy of 0.91, and (4) an available MBI-HSS app for nurses predicting BO was successfully developed and demonstrated in this study.

Conclusions

The 20-item model with the 38 parameters estimated by using CNN for improving the accuracy of nurse BO has been particularly demonstrated in Excel (Microsoft Corp). An app developed for helping nurses to self-assess job BO at an early stage is required for application in the future.

Deep neural networks in psychiatry

Machine and deep learning methods, today's core of artificial intelligence, have been applied with increasing success and impact in many commercial and research settings. They are powerful tools for large scale data analysis, prediction and classification, especially in very data-rich environments ("big data"), and have started to find their way into medical applications. Here we will first give an overview of machine learning methods, with a focus on deep and recurrent neural networks, their relation to statistics, and the core principles behind them. We will then discuss and review directions along which (deep) neural networks can be, or already have been, applied in the context of psychiatry, and will try to delineate their future potential in this area. We will also comment on an emerging area that so far has been much less well explored: by embedding semantically interpretable computational models of brain dynamics or behavior into a statistical machine learning context, insights into dysfunction beyond mere prediction and classification may be gained. Especially this marriage of computational models with statistical inference may offer insights into neural and behavioral mechanisms that could open completely novel avenues for psychiatric treatment.

Depression Prediction by Using Ecological Momentary Assessment, Actiwatch Data, and Machine Learning: Observational Study on Older Adults Living Alone

BACKGROUND

Although geriatric depression is prevalent, diagnosis using self-reporting instruments has limitations when measuring the depressed mood of older adults in a community setting. Ecological momentary assessment (EMA) by using wearable devices could be used to collect data to classify older adults into depression groups.

OBJECTIVE

The objective of this study was to develop a machine learning algorithm to predict the classification of depression groups among older adults living alone. We focused on utilizing diverse data collected through a survey, an Actiwatch, and an EMA report related to depression.

METHODS

The prediction model using machine learning was developed in 4 steps: (1) data collection, (2) data processing and representation, (3) data modeling (feature engineering and selection), and (4) training and validation to test the prediction model. Older adults (N=47), living alone in community settings, completed an EMA to report depressed moods 4 times a day for 2 weeks between May 2017 and January 2018. Participants wore an Actiwatch that measured their activity and ambient light exposure every 30 seconds for 2 weeks. At baseline and the end of the 2-week observation, depressive symptoms were assessed using the Korean versions of the Short Geriatric Depression Scale (SGDS-K) and the Hamilton Depression Rating Scale (K-HDRS). Conventional classification based on binary logistic regression was built and compared with 4 machine learning models (the logit, decision tree, boosted trees, and random forest models).

RESULTS

On the basis of the SGDS-K and K-HDRS, 38% (18/47) of the participants were classified into the probable depression group. They reported significantly lower scores of normal mood and physical activity and higher levels of white and red, green, and blue (RGB) light exposures at different degrees of various 4-hour time frames (all P<.05). Sleep efficiency was chosen for modeling through feature selection. Comparing diverse combinations of the selected variables, daily mean EMA score, daily mean activity level, white and RGB light at 4:00 pm to 8:00 pm exposure, and daily sleep efficiency were selected for modeling. Conventional classification based on binary logistic regression had a good model fit (accuracy: 0.705; precision: 0.770; specificity: 0.859; and area under receiver operating characteristic curve or AUC: 0.754). Among the 4 machine learning models, the logit model had the best fit compared with the others (accuracy: 0.910; precision: 0.929; specificity: 0.940; and AUC: 0.960).

CONCLUSIONS

This study provides preliminary evidence for developing a machine learning program to predict the classification of depression groups in older adults living alone. Clinicians should consider using this method to identify underdiagnosed subgroups and monitor daily progression regarding treatment or therapeutic intervention in the community setting. Furthermore, more efforts are needed for researchers and clinicians to diversify data collection methods by using a survey, EMA, and a sensor.

Gait can reveal sleep quality with machine learning models

Sleep quality is an important health indicator, and the current measurements of sleep rely on questionnaires, polysomnography, etc., which are intrusive, expensive or time consuming. Therefore, a more nonintrusive, inexpensive and convenient method needs to be developed. Use of the Kinect sensor to capture one’s gait pattern can reveal whether his/her sleep quality meets the requirements. Fifty-nine healthy students without disabilities were recruited as participants. The Pittsburgh Sleep Quality Index (PSQI) and Kinect sensors were used to acquire the sleep quality scores and gait data. After data preprocessing, gait features were extracted for training machine learning models that predicted sleep quality scores based on the data. The t-test indicated that the following joints had stronger weightings in the prediction: the Head, Spine Shoulder, Wrist Left, Hand Right, Thumb Left, Thumb Right, Hand Tip Left, Hip Left, and Foot Left. For sleep quality prediction, the best result was achieved by Gaussian processes, with a correlation of 0.78 (p < 0.001). For the subscales, the best result was 0.51 for daytime dysfunction (p < 0.001) by linear regression. Gait can reveal sleep quality quite well. This method is a good supplement to the existing methods in identifying sleep quality more ecologically and less intrusively.

The Extent and Coverage of Current Knowledge of Connected Health: Systematic Mapping Study

Background

This study examines the development of the connected health (CH) research landscape with a view to providing an overview of the existing CH research. The research field of CH has experienced rapid growth coinciding with increasing pressure on health care systems to become more proactive and patient centered.

Objective

This study aimed to assess the extent and coverage of the current body of knowledge in CH. In doing so, we sought to identify specific topics that have drawn the attention of CH researchers and to identify research gaps, in particular those offering opportunities for further interdisciplinary research.

Methods

A systematic mapping study that combined scientific contributions from research in the disciplines of medicine, business, computer science, and engineering was used. Overall, seven classification criteria were used to analyze the papers, including publication source, publication year, research type, empirical type, contribution type, research topic, and the medical condition studied.

Results

The search resulted in 208 papers that were analyzed by a multidisciplinary group of researchers. The results indicated a slow start for CH research but showed a more recent steady upswing since 2013. The majority of papers proposed health care solutions (77/208, 37.0%) or evaluated CH approaches (49/208, 23.5%). Case studies (59/208, 28.3%) and experiments (55/208, 26.4%) were the most popular forms of scientific validation used. Diabetes, cancer, multiple sclerosis, and heart conditions were among the most prevalent medical conditions studied.

Conclusions

We conclude that CH research has become an established field of research that has grown over the last five years. The results of this study indicate a focus on technology-driven research with a strong contribution from medicine, whereas the business aspects of CH have received less research attention.

Sleep quality prediction in caregivers using physiological signals

Most caregivers of people with dementia (CPWD) experience a high degree of stress due to the demands of providing care, especially when addressing unpredictable behavioral and psychological symptoms of dementia. Such challenging responsibilities make caregivers susceptible to poor sleep quality with detrimental effects on their overall health. Hence, monitoring caregivers' sleep quality can provide important CPWD stress assessment. Most current sleep studies are based on polysomnography, which is expensive and potentially disrupts the caregiving routine. To address these issues, we propose a clinical decision support system to predict sleep quality based on trends of physiological signals in the deep sleep stage. This system utilizes four raw physiological signals using a wearable device (E4 wristband): heart rate variability, electrodermal activity, body movement, and skin temperature. To evaluate the performance of the proposed method, analyses were conducted on a two-week period of sleep monitored on eight CPWD. The best performance is achieved using the random forest classifier with an accuracy of 75% for sleep quality, and 73% for restfulness, respectively. We found that the most important features to detect these measures are sleep efficiency (ratio of amount of time asleep to the amount of time in bed) and skin temperature. The results from our sleep analysis system demonstrate the capability of using wearable sensors to measure sleep quality and restfulness in CPWD.

Mobile Technology in Running Science and Medicine: Are We Ready?

Since the running revolution of the 1970s, one of the major challenges has been the burden of running-related injuries (RRIs). Researchers, sports medicine practitioners, and strength and conditioning coaches are striving to develop an understanding of which factors may increase an individuals risk of developing RRIs, which strategies can be used to ensure optimal rehabilitation and recovery from an injury, and how to best optimize athletic performance. This Viewpoint explores these factors to demonstrate how recent advances in mobile technology may allow us to uncover novel insights related to the science and medicine of running. J Orthop Sports Phys Ther 2019;49(3):122-125. doi:10.2519/jospt.2019.0604.

Digital Interventions for Mental Disorders: Key Features, Efficacy, and Potential for Artificial Intelligence Applications

Mental disorders are highly prevalent and often remain untreated. Many limitations of conventional face-to-face psychological interventions could potentially be overcome through Internet-based and mobile-based interventions (IMIs). This chapter introduces core features of IMIs, describes areas of application, presents evidence on the efficacy of IMIs as well as potential effect mechanisms, and delineates how Artificial Intelligence combined with IMIs may improve current practices in the prevention and treatment of mental disorders in adults. Meta-analyses of randomized controlled trials clearly show that therapist-guided IMIs can be highly effective for a broad range of mental health problems. Whether the effects of unguided IMIs are also clinically relevant, particularly under routine care conditions, is less clear. First studies on IMIs for the prevention of mental disorders have shown promising results. Despite limitations and challenges, IMIs are increasingly implemented into routine care worldwide. IMIs are also well suited for applications of Artificial Intelligence and Machine Learning, which provides ample opportunities to improve the identification and treatment of mental disorders. Together with methodological innovations, these approaches may also deepen our understanding of how psychological interventions work, and why. Ethical and professional restraints as well as potential contraindications of IMIs, however, should also be considered. In sum, IMIs have a high potential for improving the prevention and treatment of mental health disorders across various indications, settings, and populations. Therefore, implementing IMIs into routine care as both adjunct and alternative to face-to-face treatment is highly desirable. Technological advancements may further enhance the variability and flexibility of IMIs, and thus even further increase their impact in people's lives in the future.

Rhythms of life: circadian disruption and brain disorders across the lifespan

Many processes in the human body - including brain function - are regulated over the 24-hour cycle, and there are strong associations between disrupted circadian rhythms (for example, sleep-wake cycles) and disorders of the CNS. Brain disorders such as autism, depression and Parkinson disease typically develop at certain stages of life, and circadian rhythms are important during each stage of life for the regulation of processes that may influence the development of these disorders. Here, we describe circadian disruptions observed in various brain disorders throughout the human lifespan and highlight emerging evidence suggesting these disruptions affect the brain. Currently, much of the evidence linking brain disorders and circadian dysfunction is correlational, and so whether and what kind of causal relationships might exist are unclear. We therefore identify remaining questions that may direct future research towards a better understanding of the links between circadian disruption and CNS disorders.

An Activity Tracker and Its Accompanying App as a Motivator for Increased Exercise and Better Sleeping Habits for Youths in Need of Social Care: Field Study

Background

The number of mobile self-tracking devices connected to the Web has exploded in today’s society. With these wearable activity trackers related to Web 2.0 apps and social media have come new ways of monitoring, measuring, representing, and sharing experiences of the human body. New opportunities related to health and new areas of implementation for professionals have appeared, and one identified area that can benefit from mobile health technologies is social work.

Objective

There are still only a small number of papers reporting the results from studying wearable activity trackers and accompanying apps in the context of agency-based social work. This study aimed to contribute to the identified shortage by presenting results from a research project framed by the following overarching question: What effects will the studied youths in need of social care experience in relation to exercise and sleep as the result of using a wearable activity tracker and its accompanying app?

Methods

A field study framed by action research was performed. The study concerned vulnerable youths living in a Swedish municipality’s care and accommodation home that tried out an activity tracker and its accompanying app.

Results

The results from the study confirm previously published research results reporting that instant graphical feedback, sharing information, and being part of a social community can have a positive impact on lifestyle changes. In addition, this study’s main results are that (1) the most important factor for positive health-related lifestyle changes was the establishment of personal long-term goals and (2) professional social workers found the studied technology to function as a valuable counseling tool, opening up avenues for lifestyle talks that otherwise were hard to undertake.

Conclusions

This study demonstrates how an activity tracker and its accompanying app can open up a topic for discussion regarding how vulnerable youths can achieve digital support for changing unhealthy lifestyle patterns, and it shows that the technology might be a valuable counseling tool for professionals in social work.

Combining Continuous Smartphone Native Sensors Data Capture and Unsupervised Data Mining Techniques for Behavioral Changes Detection: A Case Series of the Evidence-Based Behavior (eB2) Study

BACKGROUND

The emergence of smartphones, wearable sensor technologies, and smart homes allows the nonintrusive collection of activity data. Thus, health-related events, such as activities of daily living (ADLs; eg, mobility patterns, feeding, sleeping, ...) can be captured without patients' active participation. We designed a system to detect changes in the mobility patterns based on the smartphone's native sensors and advanced machine learning and signal processing techniques.

OBJECTIVE

The principal objective of this work is to assess the feasibility of detecting mobility pattern changes in a sample of outpatients with depression using the smartphone's sensors. The proposed method processed the data acquired by the smartphone using an unsupervised detection technique.

METHODS

In this study, 38 outpatients from the Hospital Fundación Jiménez Díaz Psychiatry Department (Madrid, Spain) participated. The Evidence-Based Behavior (eB²) app was downloaded by patients on the day of recruitment and configured with the assistance of a physician. The app captured the following data: inertial sensors, physical activity, phone calls and message logs, app usage, nearby Bluetooth and Wi-Fi connections, and location. We applied a change-point detection technique to location data on a sample of 9 outpatients recruited between April 6, 2017 and December 14, 2017. The change-point detection was based only on location information, but the eB² platform allowed for an easy integration of additional data. The app remained running in the background on patients' smartphone during the study participation.

RESULTS

The principal outcome measure was the identification of mobility pattern changes based on an unsupervised detection technique applied to the smartphone's native sensors data. Here, results from 5 patients' records are presented as a case series. The eB² system detected specific mobility pattern changes according to the patients' activity, which may be used as indicators of behavioral and clinical state changes.

CONCLUSIONS

The proposed technique could automatically detect changes in the mobility patterns of outpatients who took part in this study. Assuming these mobility pattern changes correlated with behavioral changes, we have developed a technique that may identify possible relapses or clinical changes. Nevertheless, it is important to point out that the detected changes are not always related to relapses and that some clinical changes cannot be detected by the proposed method.

A P5 Approach to m-Health: Design Suggestions for Advanced Mobile Health Technology

In recent years, technology has been developed as an important resource for health care management, especially in regard to chronic conditions. In the broad field of eHealth, mobile technology (mHealth) is increasingly used to empower patients not only in disease management but also in the achievement of positive experiences and experiential growth. mHealth tools are considered powerful because, unlike more traditional Internet-based tools, they allow patients to be continuously monitored and followed by their own mobile devices and to have continual access to resources (e.g., mobile apps or functions) supporting health care management activities. However, the literature has shown that, in many cases, such technology not accepted and/or adopted in the long term by its users. To address this issue, this article reviews the main factors influencing mHealth technology acceptance/adoption in health care. Finally, based on the main aspects emerging from the review, we propose an innovative approach to mHealth design and implementation, namely P5 mHealth. Relying on the P5 approach to medicine and health care, this approach provides design suggestions to address mHealth adoption issues already at the initial stages of development of the technologies.

A Mobile Sleep-Management Learning System for Improving Students' Sleeping Habits by Integrating a Self-Regulated Learning Strategy: Randomized Controlled Trial

Background

Insomnia can significantly affect students’ learning performance. Researchers have indicated the importance and challenge of coping with insomnia using nondrug treatments, such as cognitive behavioral therapy (CBT) for insomnia. However, it is easy for the traditional CBT for insomnia to be interrupted owing to the overly lengthy period of sleep therapy. Self-regulated learning (SRL) strategies are known to be an effective approach for helping students improve their time management, as well as their ability to set learning goals and adopt learning strategies.

Objective

The objective of this study was to propose a mobile sleep-management learning system integrated with SRL strategies and CBT.

Methods

A total of 18 undergraduate students from a university in northern Taiwan participated in the 2-week experiment of using this sleep-management system.

Results

The experimental results showed that the proposed approach was useful and easy for students to use. In addition, the number of students with insomnia significantly decreased; that is, the proposed approach could help students improve their sleep quality and cultivate better sleeping habits, which is important for them to enhance their learning efficiency.

Conclusions

With the assistance of this proposed approach, students can plan their daily life by setting goals, applying strategies, monitoring their life habits process, and modifying strategies to cultivate good learning and healthy lifestyle habits.

Trial Registration

Government Research Bulletin MOST104-3011-E038-001; https://www.grb.gov.tw/search/planDetail? id=11568383&docId=467988 (Archived by WebCite at http://www.webcitation.org/73MnPHNri)

Multimodal Ambulatory Sleep Detection Using LSTM Recurrent Neural Networks

Unobtrusive and accurate ambulatory methods are needed to monitor long-term sleep patterns for improving health. Previously developed ambulatory sleep detection methods rely either in whole or in part on self-reported diary data as ground truth, which is a problem, since people often do not fill them out accurately. This paper presents an algorithm that uses multimodal data from smartphones and wearable technologies to detect sleep/wake state and sleep onset/offset using a type of recurrent neural network with long-short-term memory (LSTM) cells for synthesizing temporal information. We collected 5580 days of multimodal data from 186 participants and compared the new method for sleep/wake classification and sleep onset/offset detection to, first, nontemporal machine learning methods and, second, a state-of-the-art actigraphy software. The new LSTM method achieved a sleep/wake classification accuracy of 96.5%, and sleep onset/offset detection F₁ scores of 0.86 and 0.84, respectively, with mean absolute errors of 5.0 and 5.5 min, res-pectively, when compared with sleep/wake state and sleep onset/offset assessed using actigraphy and sleep diaries. The LSTM results were statistically superior to those from nontemporal machine learning algorithms and the actigraphy software. We show good generalization of the new algorithm by comparing participant-dependent and participant-independent models, and we show how to make the model nearly realtime with slightly reduced performance.

Risperidone Provides Better Improvement of Sleep Disturbances Than Haloperidol Therapy in Schizophrenia Patients With Cannabis-Positive Urinalysis

A high percentage of persons with Schizophrenia also uses Cannabis and this may potentially alter the therapeutic benefits of the antipsychotic medications prescribed. The aim of this study was to assess the impact of Cannabis usage on antipsychotic therapy of sleep disturbances in schizophrenia subjects. Male subjects, ≥18 years, admitted to the University Hospital of the West Indies psychiatric ward between October 2015 and October 2016, and diagnosed with schizophrenia were recruited for the study. Following written informed consent to the study, subjects were prescribed either risperidone monotherapy or haloperidol monotherapy orally for 14 days and classified as Cannabis users (CU) or non-users (non-CU), with presence/absence of Cannabis metabolite in urine samples. After 1 week of admission, subjects wore the Actiwatch-2 device, to record sleep data for 7 consecutive nights. Inferential statistical analysis involved non-parametric tests, expressed as median and inter-quartile ranges (IQR), with p<0.05 considered statistically significant. Fifty subjects were assessed, with a median (IQR) age of 28 (14) years. Majority (30; 60%) were CU, displaying longer sleep latency (SL) than non-CU when receiving haloperidol; but equivalent SL when receiving risperidone. In comparison to non-CU, the CU also displayed longer time in bed, but shorter durations asleep, awoke more frequently during the nights and for longer durations, whether receiving haloperidol or risperidone. This resulted in lower sleep efficiency for the CU (<85%) compared to the non-CU (≥85%). Over the study period, sleep efficiency was significantly improved for non-CU receiving either risperidone (p = 0.032) or haloperidol (p = 0.010); but was only significantly improved with risperidone for the CU (p = 0.045). It is apparent that the presence of Cannabis may be impacting the therapeutic benefits of antipsychotic drugs on sleep. In schizophrenia subjects with concomitant Cannabis use, risperidone is more beneficial than haloperidol in improving sleep efficiency.