Continuous Assessment of Function and Disability via Mobile Sensing: Real-World Data-Driven Feasibility Study.

BACKGROUND: Functional limitations are associated with poor clinical outcomes, higher mortality, and disability rates, especially in older adults. Continuous assessment of patients' functionality is important for clinical practice; however, traditional questionnaire-based assessment methods are very time-consuming and infrequently used. Mobile sensing offers a great range of sources that can assess function and disability daily. OBJECTIVE: This work aims to prove the feasibility of an interpretable machine learning pipeline for predicting function and disability based on the World Health Organization Disability Assessment Schedule (WHODAS) 2.0 outcomes of clinical outpatients, using passively collected digital biomarkers. METHODS: One-month-long behavioral time-series data consisting of physical and digital activity descriptor variables were summarized using statistical measures (minimum, maximum, mean, median, SD, and IQR), creating 64 features that were used for prediction. We then applied a sequential feature selection to each WHODAS 2.0 domain (cognition, mobility, self-care, getting along, life activities, and participation) in order to find the most descriptive features for each domain. Finally, we predicted the WHODAS 2.0 functional domain scores using linear regression using the best feature subsets. We reported the mean absolute errors and the mean absolute percentage errors over 4 folds as goodness-of-fit statistics to evaluate the model and allow for between-domain performance comparison. RESULTS: Our machine learning-based models for predicting patients' WHODAS functionality scores per domain achieved an average (across the 6 domains) mean absolute percentage error of 19.5%, varying between 14.86% (self-care domain) and 27.21% (life activities domain). We found that 5-19 features were sufficient for each domain, and the most relevant being the distance traveled, time spent at home, time spent walking, exercise time, and vehicle time. CONCLUSIONS: Our findings show the feasibility of using machine learning-based methods to assess functional health solely from passively sensed mobile data. The feature selection step provides a set of interpretable features for each domain, ensuring better explainability to the models' decisions-an important aspect in clinical practice.

Automatic patient functionality assessment from multimodal data using deep learning techniques - Development and feasibility evaluation.

Wearable devices and mobile sensors enable the real-time collection of an abundant source of physiological and behavioural data unobtrusively. Unlike traditional in-person evaluation or ecological momentary assessment (EMA) questionnaire-based approaches, these data sources open many possibilities in remote patient monitoring. However, defining robust models is challenging due to the data's noisy and frequently missing observations. This work proposes an attention-based Long Short-Term Memory (LSTM) neural network-based pipeline for predicting mobility impairment based on WHODAS 2.0 evaluation from such digital biomarkers. Furthermore, we addressed the missing observation problem by utilising hidden Markov models and the possibility of including information from unlabelled samples via transfer learning. We validated our approach using two wearable/mobile sensor data sets collected in the wild and socio-demographic information about the patients. Our results showed that in the WHODAS 2.0 mobility impairment prediction task, the proposed pipeline outperformed a prior baseline while additionally providing interpretability with attention heatmaps. Moreover, using a much smaller cohort via task transfer learning, the same model could learn to predict generalised anxiety severity accurately based on GAD-7 scores.

Smartphone-based Ecological Momentary Intervention for secondary prevention of suicidal thoughts and behaviour: protocol for the SmartCrisis V.2.0 randomised clinical trial.

INTRODUCTION: Suicide is one of the leading public health issues worldwide. Mobile health can help us to combat suicide through monitoring and treatment. The SmartCrisis V.2.0 randomised clinical trial aims to evaluate the effectiveness of a smartphone-based Ecological Momentary Intervention to prevent suicidal thoughts and behaviour. METHODS AND ANALYSIS: The SmartCrisis V.2.0 study is a randomised clinical trial with two parallel groups, conducted among patients with a history of suicidal behaviour treated at five sites in France and Spain. The intervention group will be monitored using Ecological Momentary Assessment (EMA) and will receive an Ecological Momentary Intervention called 'SmartSafe' in addition to their treatment as usual (TAU). TAU will consist of mental health follow-up of the patient (scheduled appointments with a psychiatrist) in an outpatient Suicide Prevention programme, with predetermined clinical appointments according to the Brief Intervention Contact recommendations (1, 2, 4, 7 and 11 weeks and 4, 6, 9 and 12 months). The control group would receive TAU and be monitored using EMA. ETHICS AND DISSEMINATION: This study has been approved by the Ethics Committee of the University Hospital Fundación Jiménez Díaz. It is expected that, in the near future, our mobile health intervention and monitoring system can be implemented in routine clinical practice. Results will be disseminated through peer-reviewed journals and psychiatric congresses. Reference number EC005-21_FJD. Participants gave informed consent to participate in the study before taking part. TRIAL REGISTRATION NUMBER: NCT04775160.

Virtual reality and speech analysis for the assessment of impulsivity and decision-making: protocol for a comparison with neuropsychological tasks and self-administered questionnaires.

INTRODUCTION: Impulsivity is present in a range of mental disorders and has been associated with suicide. Traditional measures of impulsivity have certain limitations, such as the lack of ecological validity. Virtual reality (VR) may overcome these issues. This study aims to validate the VR assessment tool 'Spheres & Shield Maze Task' and speech analysis by comparing them with traditional measures. We hypothesise that these innovative tools will be reliable and acceptable by patients, potentially improving the simultaneous assessment of impulsivity and decision-making. METHODS AND ANALYSIS: This study will be carried out at the University Hospital Fundación Jiménez Díaz (Madrid, Spain). Our sample will consist of adults divided into three groups: psychiatric outpatients with a history of suicidal thoughts and/or behaviours, psychiatric outpatients without such a history and healthy volunteers. The target sample size was established at 300 participants (100 per group). Participants will complete the Barratt Impulsiveness Scale 11; the Urgency, Premeditation, Perseverance, Sensation Seeking, Positive Urgency, Impulsive Behaviour Scale; Iowa Gambling Task; Continuous Performance Test; Stop signal Task, and Go/no-go task, three questions of emotional affect, the Spheres & Shield Maze Task and two satisfaction surveys. During these tasks, participant speech will be recorded. Construct validity of the VR environment will be calculated. We will also explore the association between VR-assessed impulsivity and history of suicidal thoughts and/or behaviour, and the association between speech and impulsivity and decision-making. ETHICS AND DISSEMINATION: This study was approved by the Ethics Committee of the University Hospital Fundación Jiménez Díaz (PIC128-21_FJD). Participants will be required to provide written informed consent. The findings will be presented in a series of manuscripts that will be submitted to peer-reviewed journals for publication. TRIAL REGISTRATION NUMBER: NCT05109845; Pre-results.

Shift in Social Media App Usage During COVID-19 Lockdown and Clinical Anxiety Symptoms: Machine Learning-Based Ecological Momentary Assessment Study.

BACKGROUND: Anxiety symptoms during public health crises are associated with adverse psychiatric outcomes and impaired health decision-making. The interaction between real-time social media use patterns and clinical anxiety during infectious disease outbreaks is underexplored. OBJECTIVE: We aimed to evaluate the usage pattern of 2 types of social media apps (communication and social networking) among patients in outpatient psychiatric treatment during the COVID-19 surge and lockdown in Madrid, Spain and their short-term anxiety symptoms (7-item General Anxiety Disorder scale) at clinical follow-up. METHODS: The individual-level shifts in median social media usage behavior from February 1 through May 3, 2020 were summarized using repeated measures analysis of variance that accounted for the fixed effects of the lockdown (prelockdown versus postlockdown), group (clinical anxiety group versus nonclinical anxiety group), the interaction of lockdown and group, and random effects of users. A machine learning-based approach that combined a hidden Markov model and logistic regression was applied to predict clinical anxiety (n=44) and nonclinical anxiety (n=51), based on longitudinal time-series data that comprised communication and social networking app usage (in seconds) as well as anxiety-associated clinical survey variables, including the presence of an essential worker in the household, worries about life instability, changes in social interaction frequency during the lockdown, cohabitation status, and health status. RESULTS: Individual-level analysis of daily social media usage showed that the increase in communication app usage from prelockdown to lockdown period was significantly smaller in the clinical anxiety group than that in the nonclinical anxiety group (F1,72=3.84, P=.05). The machine learning model achieved a mean accuracy of 62.30% (SD 16%) and area under the receiver operating curve 0.70 (SD 0.19) in 10-fold cross-validation in identifying the clinical anxiety group. CONCLUSIONS: Patients who reported severe anxiety symptoms were less active in communication apps after the mandated lockdown and more engaged in social networking apps in the overall period, which suggested that there was a different pattern of digital social behavior for adapting to the crisis. Predictive modeling using digital biomarkers-passive-sensing of shifts in category-based social media app usage during the lockdown-can identify individuals at risk for psychiatric sequelae.

Predicting Emotional States Using Behavioral Markers Derived From Passively Sensed Data: Data-Driven Machine Learning Approach.

BACKGROUND: Mental health disorders affect multiple aspects of patients' lives, including mood, cognition, and behavior. eHealth and mobile health (mHealth) technologies enable rich sets of information to be collected noninvasively, representing a promising opportunity to construct behavioral markers of mental health. Combining such data with self-reported information about psychological symptoms may provide a more comprehensive and contextualized view of a patient's mental state than questionnaire data alone. However, mobile sensed data are usually noisy and incomplete, with significant amounts of missing observations. Therefore, recognizing the clinical potential of mHealth tools depends critically on developing methods to cope with such data issues. OBJECTIVE: This study aims to present a machine learning-based approach for emotional state prediction that uses passively collected data from mobile phones and wearable devices and self-reported emotions. The proposed methods must cope with high-dimensional and heterogeneous time-series data with a large percentage of missing observations. METHODS: Passively sensed behavior and self-reported emotional state data from a cohort of 943 individuals (outpatients recruited from community clinics) were available for analysis. All patients had at least 30 days' worth of naturally occurring behavior observations, including information about physical activity, geolocation, sleep, and smartphone app use. These regularly sampled but frequently missing and heterogeneous time series were analyzed with the following probabilistic latent variable models for data averaging and feature extraction: mixture model (MM) and hidden Markov model (HMM). The extracted features were then combined with a classifier to predict emotional state. A variety of classical machine learning methods and recurrent neural networks were compared. Finally, a personalized Bayesian model was proposed to improve performance by considering the individual differences in the data and applying a different classifier bias term for each patient. RESULTS: Probabilistic generative models proved to be good preprocessing and feature extractor tools for data with large percentages of missing observations. Models that took into account the posterior probabilities of the MM and HMM latent states outperformed those that did not by more than 20%, suggesting that the underlying behavioral patterns identified were meaningful for individuals' overall emotional state. The best performing generalized models achieved a 0.81 area under the curve of the receiver operating characteristic and 0.71 area under the precision-recall curve when predicting self-reported emotional valence from behavior in held-out test data. Moreover, the proposed personalized models demonstrated that accounting for individual differences through a simple hierarchical model can substantially improve emotional state prediction performance without relying on previous days' data. CONCLUSIONS: These findings demonstrate the feasibility of designing machine learning models for predicting emotional states from mobile sensing data capable of dealing with heterogeneous data with large numbers of missing observations. Such models may represent valuable tools for clinicians to monitor patients' mood states.

Psychiatric Profiles of eHealth Users Evaluated Using Data Mining Techniques: Cohort Study.

BACKGROUND: New technologies are changing access to medical records and the relationship between physicians and patients. Professionals can now use e-mental health tools to provide prompt and personalized responses to patients with mental illness. However, there is a lack of knowledge about the digital phenotypes of patients who use e-mental health apps. OBJECTIVE: This study aimed to reveal the profiles of users of a mental health app through machine learning techniques. METHODS: We applied a nonparametric model, the Sparse Poisson Factorization Model, to discover latent features in the response patterns of 2254 psychiatric outpatients to a short self-assessment on general health. The assessment was completed through a mental health app after the first login. RESULTS: The results showed the following four different profiles of patients: (1) all patients had feelings of worthlessness, aggressiveness, and suicidal ideas; (2) one in four reported low energy and difficulties to cope with problems; (3) less than a quarter described depressive symptoms with extremely high scores in suicidal thoughts and aggressiveness; and (4) a small number, possibly with the most severe conditions, reported a combination of all these features. CONCLUSIONS: User profiles did not overlap with clinician-made diagnoses. Since each profile seems to be associated with a different level of severity, the profiles could be useful for the prediction of behavioral risks among users of e-mental health apps.

Universal mental health screening with a focus on suicidal behaviour using smartphones in a Mexican rural community: protocol for the SMART-SCREEN population-based survey.

INTRODUCTION: Mental disorders represent the second cause of years lived with disability worldwide. Suicide mortality has been targeted as a key public health concern by the WHO. Smartphone technology provides a huge potential to develop massive and fast surveys. Given the vast cultural diversity of Mexico and its abrupt orography, smartphone-based resources are invaluable in order to adequately manage resources, services and preventive measures in the population. The objective of this study is to conduct a universal suicide risk screening in a rural area of Mexico, measuring also other mental health outcomes such as depression, anxiety and alcohol and substance use disorders. METHODS AND ANALYSIS: A population-based cross-sectional study with a temporary sampling space of 9 months will be performed between September 2019 and June 2020. We expect to recruit a large percentage of the target population (at least 70%) in a short-term survey of Milpa Alta Delegation, which accounts for 137 927 inhabitants in a territorial extension of 288 km2.They will be recruited via an institutional call and a massive public campaign to fill in an online questionnaire through mobile-assisted or computer-assisted web app. This questionnaire will include data on general health, validated questionnaires including Well-being Index 5, Patient Health Questionnaire-9, Generalized Anxiety Disorder Scale 2, Alcohol Use Disorders Identification Test, selected questions of the Drug Abuse Screening Test and Columbia-Suicide Severity Rating Scales and Diagnostic and statistical manual of mental disorders (DSM-5) questions about self-harm.We will take into account information regarding time to mobile app response and geo-spatial location, and aggregated data on social, demographical and environmental variables. Traditional regression modelling, multilevel mixed methods and data-driven machine learning approaches will be used to test hypotheses regarding suicide risk factors at the individual and the population level. ETHICS AND DISSEMINATION: Ethical approval (002/2019) was granted by the Ethics Review Board of the Hospital Psiquiátrico Yucatán, Yucatán (Mexico). This protocol has been registered in ClinicalTrials.gov. The starting date of the study is 3 September 2019. Results will serve for the planning and healthcare of groups with greater mental health needs and will be disseminated via publications in peer-reviewed journal and presented at relevant mental health conferences. TRIAL REGISTRATION NUMBER: NCT04067063.

Validation of Fitbit Charge 2 and Fitbit Alta HR Against Polysomnography for Assessing Sleep in Adults With Obstructive Sleep Apnea.

STUDY OBJECTIVES: Consumer wearable devices may be a helpful method of assessing sleep, but validation is required for their use in clinical practice. Our aim was to validate two models of Fitbit sleep trackers that rely on both accelerometer and heart rate sensors against polysomnography in participants with obstructive sleep apnea (OSA). METHODS: Participants were adults presenting with symptoms of OSA and attending our outpatient sleep clinic. A polysomnography (PSG) was applied to all participants at the same time they were wearing a Fitbit sleep tracker. Using paired t tests and Bland-Altman plots, we compared the sleep measures provided by the wearable devices with those obtained by PSG. Since Fitbit devices' automatic detection of sleep start time can cause bias, we performed a correction using Huber loss function-based linear regression and a leave-one-out strategy. RESULTS: Our sample consisted of 65 patients. Diagnosis of OSA was confirmed on 55 (84.6%). There were statistically significant differences between PSG and Fitbit measures for all sleep outcomes but rapid eye movement sleep. Fitbit devices overestimated total sleep time, and underestimated wake after sleep onset and sleep onset latency. After correction of bias, Fitbit-delivered measures of sleep onset latency did not significantly differ of those provided by PSG. CONCLUSIONS: Fitbit wearable devices showed an acceptable sensitivity but poor specificity. Consumer sleep trackers still have insufficient accuracy for clinical settings, especially in clinical populations. Solving technical issues and optimizing clinically-oriented features could make them apt for their use in clinical practice in a nondistant future.

A configurable and low-power mixed signal SoC for portable ECG monitoring applications.

This paper describes a mixed-signal ECG System-on-Chip (SoC) that is capable of implementing configurable functionality with low-power consumption for portable ECG monitoring applications. A low-voltage and high performance analog front-end extracts 3-channel ECG signals and single channel electrode-tissue-impedance (ETI) measurement with high signal quality. This can be used to evaluate the quality of the ECG measurement and to filter motion artifacts. A custom digital signal processor consisting of 4-way SIMD processor provides the configurability and advanced functionality like motion artifact removal and R peak detection. A built-in 12-bit analog-to-digital converter (ADC) is capable of adaptive sampling achieving a compression ratio of up to 7, and loop buffer integration reduces the power consumption for on-chip memory access. The SoC is implemented in 0.18 µm CMOS process and consumes 32 µ W from a 1.2 V while heart beat detection application is running, and integrated in a wireless ECG monitoring system with Bluetooth protocol. Thanks to the ECG SoC, the overall system power consumption can be reduced significantly.

Power impact of loop buffer schemes for biomedical wireless sensor nodes.

Instruction memory organisations are pointed out as one of the major sources of energy consumption in embedded systems. As these systems are characterised by restrictive resources and a low-energy budget, any enhancement in this component allows not only to decrease the energy consumption but also to have a better distribution of the energy budget throughout the system. Loop buffering is an effective scheme to reduce energy consumption in instruction memory organisations. In this paper, the loop buffer concept is applied in real-life embedded applications that are widely used in biomedical Wireless Sensor Nodes, to show which scheme of loop buffer is more suitable for applications with certain behaviour. Post-layout simulations demonstrate that a trade-off exists between the complexity of the loop buffer architecture and the energy savings of utilising it. Therefore, the use of loop buffer architectures in order to optimise the instruction memory organisation from the energy efficiency point of view should be evaluated carefully, taking into account two factors: (1) the percentage of the execution time of the application that is related to the execution of the loops, and (2) the distribution of the execution time percentage over each one of the loops that form the application.

Differentiation of ventricular and supraventricular tachycardias based on the analysis of the first postpacing interval after sequential anti-tachycardia pacing in implantable cardioverter-defibrillator patients.

BACKGROUND: Current discrimination algorithms do not completely avoid inappropriate tachycardia detection. OBJECTIVES: This study analyzes the discrimination capability of the changes of the first postpacing interval (FPPI) after successive bursts of anti-tachycardia pacing (ATP) trains in implantable cardioverter-defibrillator (ICD)-recorded tachycardias. METHODS: We included 50 ICD patients in this prospective study. We hypothesized that the FPPI variability (FPPIV) when comparing bursts with different numbers of beats would be shorter in ventricular tachycardias (VTs) compared with supraventricular tachycardias (SVTs). The ATP (5-10 pulses, 91% of tachycardia cycle length) was programmed for tachycardias >240 ms. RESULTS: Anti-tachycardia pacing was delivered during 37 sinus tachycardias (STs) in an exercise test, 96 induced VTs in an electrophysiological study, and 198 spontaneous episodes (144 VTs and 54 SVTs). The FPPI remained stable after all ATP bursts in VT but changed continuously in SVT; when comparing bursts of 5 and 10 pulses, the FPPIV was shorter in VT (34 +/- 65 vs.138 +/- 69, P<.0001, in all T and 12 +/- 20 vs. 138 +/- 69, P<.0001, in T>or=320 ms) than in SVT. In T>or=320 ms an FPPIV