Passive sensing data predicts stress in university students: a supervised machine learning method for digital phenotyping.

Introduction: University students are particularly susceptible to developing high levels of stress, which occur when environmental demands outweigh an individual's ability to cope. The growing advent of mental health smartphone apps has led to a surge in use by university students seeking ways to help them cope with stress. Use of these apps has afforded researchers the unique ability to collect extensive amounts of passive sensing data including GPS and step detection. Despite this, little is known about the relationship between passive sensing data and stress. Further, there are no established methodologies or tools to predict stress from passive sensing data in this group. Methods: In this study, we establish a clear machine learning-based methodological pipeline for processing passive sensing data and extracting features that may be relevant in the context of mental health. Results: We then use this methodology to determine the relationship between passive sensing data and stress in university students. Discussion: In doing so, we offer the first proof-of-principle data for the utility of our methodological pipeline and highlight that passive sensing data can indeed digitally phenotype stress in university students. Clinical trial registration: Australia New Zealand Clinical Trials Registry (ANZCTR), identifier ACTRN12621001223820.

MoodCapture: Depression Detection Using In-the-Wild Smartphone Images.

MoodCapture presents a novel approach that assesses depression based on images automatically captured from the front-facing camera of smartphones as people go about their daily lives. We collect over 125,000 photos in the wild from N=177 participants diagnosed with major depressive disorder for 90 days. Images are captured naturalistically while participants respond to the PHQ-8 depression survey question: "I have felt down, depressed, or hopeless". Our analysis explores important image attributes, such as angle, dominant colors, location, objects, and lighting. We show that a random forest trained with face landmarks can classify samples as depressed or non-depressed and predict raw PHQ-8 scores effectively. Our post-hoc analysis provides several insights through an ablation study, feature importance analysis, and bias assessment. Importantly, we evaluate user concerns about using MoodCapture to detect depression based on sharing photos, providing critical insights into privacy concerns that inform the future design of in-the-wild image-based mental health assessment tools.

Passively sensing smartphone use in teens with rates of use by sex and across operating systems.

Youth screen media activity is a growing concern, though few studies include objective usage data. Through the longitudinal, U.S.-based Adolescent Brain Cognitive Development (ABCD) Study, youth (mage = 14; n = 1415) self-reported their typical smartphone use and passively recorded three weeks of smartphone use via the ABCD-specific Effortless Assessment Research System (EARS) application. Here we describe and validate passively-sensed smartphone keyboard and app use measures, provide code to harmonize measures across operating systems, and describe trends in adolescent smartphone use. Keyboard and app-use measures were reliable and positively correlated with one another (r = 0.33) and with self-reported use (rs = 0.21-0.35). Participants recorded a mean of 5 h of daily smartphone use, which is two more hours than they self-reported. Further, females logged more smartphone use than males. Smartphone use was recorded at all hours, peaking on average from 8 to 10 PM and lowest from 3 to 5 AM. Social media and texting apps comprised nearly half of all use. Data are openly available to approved investigators ( https://nda.nih.gov/abcd/ ). Information herein can inform use of the ABCD dataset to longitudinally study health and neurodevelopmental correlates of adolescent smartphone use.

The Fully Understanding Eating and Lifestyle Behaviors (FUEL) trial: Protocol for a cohort study harnessing digital health tools to phenotype dietary non-adherence behaviors during lifestyle intervention.

Objective: Lifestyle intervention can produce clinically significant weight loss and reduced disease risk/severity for many individuals with overweight/obesity. Dietary lapses, instances of non-adherence to the recommended dietary goal(s) in lifestyle intervention, are associated with less weight loss and higher energy intake. There are distinct "types" of dietary lapse (e.g., eating an off-plan food, eating a larger portion), and behavioral, psychosocial, and contextual mechanisms may differ across dietary lapse types. Some lapse types also appear to impact weight more than others. Elucidating clear lapse types thus has potential for understanding and improving adherence to lifestyle intervention. Methods: This 18-month observational cohort study will use real-time digital assessment tools within a multi-level factor analysis framework to uncover "lapse phenotypes" and understand their impact on clinical outcomes. Adults with overweight/obesity (n = 150) will participate in a 12-month online lifestyle intervention and 6-month weight loss maintenance period. Participants will complete 14-day lapse phenotyping assessment periods at baseline, 3, 6, 12, and 18 months in which smartphone surveys, wearable devices, and geolocation will assess dietary lapses and relevant phenotyping characteristics. Energy intake (via 24-h dietary recall) and weight will be collected at each assessment period. Results: This trial is ongoing; data collection began on 31 October 2022 and is scheduled to complete by February 2027. Conclusion: Results will inform novel precision tools to improve dietary adherence in lifestyle intervention, and support updated theoretical models of adherence behavior. Additionally, these phenotyping methods can likely be leveraged to better understand non-adherence to other health behavior interventions. Trial Registration: This study was prospectively registered https://clinicaltrials.gov/study/NCT05562427.

A New Method for Detecting Dehydration of the Human Body Using Non-Contact Millimeter Wave Radiometry.

Dehydration is a common problem in the aging population. Medical professionals can detect dehydration using either blood or urine tests. This requires experimental tests in the lab as well as urine and blood samples to be obtained from the patients. This paper proposed 100 GHz millimeter wave radiometry for early detection of dehydration. Reflectance measurements were performed on healthy and dehydrated patients of both genders (120 males and 80 females) in the aging population. Based on the cause of dehydration, the patient groups were divided into three categories: (1) patients dehydrated due to less thirst sensation, (2) patients dehydrated due to illnesses (vomiting and diarrhea), and (3) patients dehydrated due to diabetes. Reflectance measurements were performed on eight locations: (1) the palm, (2) the back of the hand, (3) the fingers, (4) the inner wrist, (5) the outer wrist, (6) the volar side of the arm, (7) the dorsal surface of the arm, and (8) the elbow. Skin dehydrated due to vomiting and diarrhea was found to have lower reflectance at all the measurement locations compared with healthy and other types of dehydrated skin. The elbow region showed the highest difference in reflectance between healthy and dehydrated skin. This indicates that radiometric sensitivity is sufficient to detect dehydration in a few seconds. This will reduce the patient's waiting time and the healthcare professional's intervention time as well as allow early treatment of dehydration, thus avoiding admission to hospitals.

Contextual AI Journaling: Integrating LLM and Time Series Behavioral Sensing Technology to Promote Self-Reflection and Well-being using the MindScape App.

MindScape aims to study the benefits of integrating time series behavioral patterns (e.g., conversational engagement, sleep, location) with Large Language Models (LLMs) to create a new form of contextual AI journaling, promoting self-reflection and well-being. We argue that integrating behavioral sensing in LLMs will likely lead to a new frontier in AI. In this Late-Breaking Work paper, we discuss the MindScape contextual journal App design that uses LLMs and behavioral sensing to generate contextual and personalized journaling prompts crafted to encourage self-reflection and emotional development. We also discuss the MindScape study of college students based on a preliminary user study and our upcoming study to assess the effectiveness of contextual AI journaling in promoting better well-being on college campuses. MindScape represents a new application class that embeds behavioral intelligence in AI.

Detecting and Understanding Social Influence During Drinking Situations: Protocol for a Bluetooth-Based Sensor Feasibility and Acceptability Study.

BACKGROUND: High-risk alcohol consumption among young adults frequently occurs in the presence of peers who are also drinking. A high-risk drinking situation may consist of particular social network members who have a primary association with drinking. Fine-grained approaches such as ecological momentary assessment (EMA) are growing in popularity for studying real-time social influence, but studies using these approaches exclusively rely on participant self-report. Passive indicators of peer presence using Bluetooth-based technology to detect real-time interactions have the potential to assist in the development of just-in-time interventions. OBJECTIVE: This study seeks to examine the feasibility and acceptability of using a Bluetooth-based sensor and smartphone app to measure social contact in real-world drinking situations. METHODS: Young adults (N=20) who drink heavily and report social drinking will be recruited from the community to participate in a 3-week EMA study. Using a social network interview, index participants will identify and recruit 3 of their friends to carry a Bluetooth beacon. Participants will complete a series of EMA reports on their own personal Android devices including random reports; morning reports; first-drink reports; and signal-contingent reports, which are triggered following the detection of a beacon carried by a peer participant. EMA will assess alcohol use and characteristics of the social environment, including who is nearby and who is drinking. For items about peer proximity and peer drinking, a customized peer list will be presented to participants. Feedback about the study protocol will be ascertained through weekly contact with both index and peer participants, followed by a qualitative interview at the end of the study. We will examine the feasibility and acceptability of recruitment, enrollment of participants and peers, and retention. Feasibility will be determined using indexes of eligibility, enrollment, and recruitment. Acceptability will be determined through participant enrollment and retention, protocol compliance, and participant-reported measures of acceptability. Feasibility and acceptability for peer participants will be informed by enrollment rates, latency to enrollment, compliance with carrying the beacon, and self-reported reasons for compliance or noncompliance with beacon procedures. Finally, EMA data about peer proximity and peer drinking will support the validity of the peer selection process. RESULTS: Participant recruitment began in February 2023, and enrollment was completed in December 2023. Results will be reported in 2025. CONCLUSIONS: The protocol allows us to examine the feasibility and acceptability of a Bluetooth-based sensor for the detection of social contact between index participants and their friends, including social interactions during real-world drinking situations. Data from this study will inform just-in-time adaptive interventions seeking to address drinking in the natural environment by providing personalized feedback about a high-risk social context and alerting an individual that they are in a potentially unsafe situation. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/50650.

Framework for Ranking Machine Learning Predictions of Limited, Multimodal, and Longitudinal Behavioral Passive Sensing Data: Combining User-Agnostic and Personalized Modeling.

BACKGROUND: Passive mobile sensing provides opportunities for measuring and monitoring health status in the wild and outside of clinics. However, longitudinal, multimodal mobile sensor data can be small, noisy, and incomplete. This makes processing, modeling, and prediction of these data challenging. The small size of the data set restricts it from being modeled using complex deep learning networks. The current state of the art (SOTA) tackles small sensor data sets following a singular modeling paradigm based on traditional machine learning (ML) algorithms. These opt for either a user-agnostic modeling approach, making the model susceptible to a larger degree of noise, or a personalized approach, where training on individual data alludes to a more limited data set, giving rise to overfitting, therefore, ultimately, having to seek a trade-off by choosing 1 of the 2 modeling approaches to reach predictions. OBJECTIVE: The objective of this study was to filter, rank, and output the best predictions for small, multimodal, longitudinal sensor data using a framework that is designed to tackle data sets that are limited in size (particularly targeting health studies that use passive multimodal sensors) and that combines both user agnostic and personalized approaches, along with a combination of ranking strategies to filter predictions. METHODS: In this paper, we introduced a novel ranking framework for longitudinal multimodal sensors (FLMS) to address challenges encountered in health studies involving passive multimodal sensors. Using the FLMS, we (1) built a tensor-based aggregation and ranking strategy for final interpretation, (2) processed various combinations of sensor fusions, and (3) balanced user-agnostic and personalized modeling approaches with appropriate cross-validation strategies. The performance of the FLMS was validated with the help of a real data set of adolescents diagnosed with major depressive disorder for the prediction of change in depression in the adolescent participants. RESULTS: Predictions output by the proposed FLMS achieved a 7% increase in accuracy and a 13% increase in recall for the real data set. Experiments with existing SOTA ML algorithms showed an 11% increase in accuracy for the depression data set and how overfitting and sparsity were handled. CONCLUSIONS: The FLMS aims to fill the gap that currently exists when modeling passive sensor data with a small number of data points. It achieves this through leveraging both user-agnostic and personalized modeling techniques in tandem with an effective ranking strategy to filter predictions.

Digital Phenotyping for Stress, Anxiety, and Mild Depression: Systematic Literature Review.

BACKGROUND: Unaddressed early-stage mental health issues, including stress, anxiety, and mild depression, can become a burden for individuals in the long term. Digital phenotyping involves capturing continuous behavioral data via digital smartphone devices to monitor human behavior and can potentially identify milder symptoms before they become serious. OBJECTIVE: This systematic literature review aimed to answer the following questions: (1) what is the evidence of the effectiveness of digital phenotyping using smartphones in identifying behavioral patterns related to stress, anxiety, and mild depression? and (2) in particular, which smartphone sensors are found to be effective, and what are the associated challenges? METHODS: We used the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) process to identify 36 papers (reporting on 40 studies) to assess the key smartphone sensors related to stress, anxiety, and mild depression. We excluded studies conducted with nonadult participants (eg, teenagers and children) and clinical populations, as well as personality measurement and phobia studies. As we focused on the effectiveness of digital phenotyping using smartphones, results related to wearable devices were excluded. RESULTS: We categorized the studies into 3 major groups based on the recruited participants: studies with students enrolled in universities, studies with adults who were unaffiliated to any particular organization, and studies with employees employed in an organization. The study length varied from 10 days to 3 years. A range of passive sensors were used in the studies, including GPS, Bluetooth, accelerometer, microphone, illuminance, gyroscope, and Wi-Fi. These were used to assess locations visited; mobility; speech patterns; phone use, such as screen checking; time spent in bed; physical activity; sleep; and aspects of social interactions, such as the number of interactions and response time. Of the 40 included studies, 31 (78%) used machine learning models for prediction; most others (n=8, 20%) used descriptive statistics. Students and adults who experienced stress, anxiety, or depression visited fewer locations, were more sedentary, had irregular sleep, and accrued increased phone use. In contrast to students and adults, less mobility was seen as positive for employees because less mobility in workplaces was associated with higher performance. Overall, travel, physical activity, sleep, social interaction, and phone use were related to stress, anxiety, and mild depression. CONCLUSIONS: This study focused on understanding whether smartphone sensors can be effectively used to detect behavioral patterns associated with stress, anxiety, and mild depression in nonclinical participants. The reviewed studies provided evidence that smartphone sensors are effective in identifying behavioral patterns associated with stress, anxiety, and mild depression.

Optimizing Outcomes in Psychotherapy for Anxiety Disorders Using Smartphone-Based and Passive Sensing Features: Protocol for a Randomized Controlled Trial.

BACKGROUND: Psychotherapies, such as cognitive behavioral therapy (CBT), currently have the strongest evidence of durable symptom changes for most psychological disorders, such as anxiety disorders. Nevertheless, only about half of individuals treated with CBT benefit from it. Predictive algorithms, including digital assessments and passive sensing features, could better identify patients who would benefit from CBT, and thus, improve treatment choices. OBJECTIVE: This study aims to establish predictive features that forecast responses to transdiagnostic CBT in anxiety disorders and to investigate key mechanisms underlying treatment responses. METHODS: This study is a 2-armed randomized controlled clinical trial. We include patients with anxiety disorders who are randomized to either a transdiagnostic CBT group or a waitlist (referred to as WAIT). We index key features to predict responses prior to starting treatment using subjective self-report questionnaires, experimental tasks, biological samples, ecological momentary assessments, activity tracking, and smartphone-based passive sensing to derive a multimodal feature set for predictive modeling. Additional assessments take place weekly at mid- and posttreatment and at 6- and 12-month follow-ups to index anxiety and depression symptom severity. We aim to include 150 patients, randomized to CBT versus WAIT at a 3:1 ratio. The data set will be subject to full feature and important features selected by minimal redundancy and maximal relevance feature selection and then fed into machine leaning models, including eXtreme gradient boosting, pattern recognition network, and k-nearest neighbors to forecast treatment response. The performance of the developed models will be evaluated. In addition to predictive modeling, we will test specific mechanistic hypotheses (eg, association between self-efficacy, daily symptoms obtained using ecological momentary assessments, and treatment response) to elucidate mechanisms underlying treatment response. RESULTS: The trial is now completed. It was approved by the Cantonal Ethics Committee, Zurich. The results will be disseminated through publications in scientific peer-reviewed journals and conference presentations. CONCLUSIONS: The aim of this trial is to improve current CBT treatment by precise forecasting of treatment response and by understanding and potentially augmenting underpinning mechanisms and personalizing treatment. TRIAL REGISTRATION: ClinicalTrials.gov NCT03945617; https://clinicaltrials.gov/ct2/show/results/NCT03945617. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/42547.

Digital biomarkers in depression: A systematic review and call for standardization and harmonization of feature engineering.

BACKGROUND: General physicians misclassify depression in more than half of the cases. Researchers have explored the feasibility of leveraging passively collected data points, also called digital biomarkers, to provide more granular understanding of depression phenotypes as well as a more objective assessment of disease. METHOD: This paper provides a systematic review following the PRISMA guidelines (Page et al., 2021) to understand which digital biomarkers might be relevant for passive screening of depression. Pubmed and PsycInfo were systematically searched for studies published from 2019 to early 2024, resulting in 161 records assessed for eligibility. Excluded were intervention studies, studies focusing on a different disease or those with a lack of passive data collection. 74 studies remained for a quality assessment, after which 27 studies were included. RESULTS: The review shows that depressed participants' real-life behavior such as reduced communication with others can be tracked by passive data. Machine learning models for the classification of depression have shown accuracies up to 0.98, surpassing the quality of many standardized assessment methods. LIMITATIONS: Inconsistency of outcome reporting of current studies does not allow for drawing statistical conclusions regarding effectiveness of individual included features. The Covid-19 pandemic might have impacted the ongoing studies between 2020 and 2022. CONCLUSION: While digital biomarkers allow real-life tracking of participant's behavior and symptoms, further work is required to align the feature engineering of digital biomarkers. With shown high accuracies of assessments, connecting digital biomarkers with clinical practice can be a promising method of detecting symptoms of depression automatically.

A Bluetooth-Based Smartphone App for Detecting Peer Proximity: Protocol for Evaluating Functionality and Validity.

BACKGROUND: While ecological momentary assessment (EMA) is commonly used to study social contexts and social influence in the real world, EMA almost exclusively relies on participant self-report of present circumstances, including the proximity to influential peers. There is the potential for developing a proximity sensing approach that uses small Bluetooth beacons and smartphone-based detection and data collection to collect information about interactions between individuals passively in real time. OBJECTIVE: This paper aims to describe the methods for evaluating the functionality and validity of a Bluetooth-based beacon and a smartphone app to identify when ≥2 individuals are physically proximal. METHODS: We will recruit 20 participants aged 18 to 29 years with Android smartphones to complete a 3-week study during which beacon detection and self-report data will be collected using a smartphone app (MEI Research). Using an interviewer-administered social network interview, participants will identify up to 3 peers of the same age who are influential on health behavior (alcohol use in this study). These peers will be asked to carry a Bluetooth beacon (Kontakt asset tag) for the duration of the study; each beacon has a unique ID that, when detected, will be recorded by the app on the participant's phone. Participants will be prompted to respond to EMA surveys (signal-contingent reports) when a peer beacon encounter meets our criteria and randomly 3 times daily (random reports) and every morning (morning reports) to collect information about the presence of peers. In all reports, the individualized list of peers will be presented to participants, followed by questions about peer and participant behavior, including alcohol use. Data from multiple app data sets, including beacon encounter specifications, notification, and app logs, participant EMA self-reports and postparticipation interviews, and peer surveys, will be used to evaluate project goals. We will examine the functionality of the technology, including the stability of the app (eg, app crashes and issues opening the app), beacon-to-app detection (ie, does the app detect proximal beacons?), and beacon encounter notification when encounter criteria are met. The validity of the technology will be defined as the concordance between passive detection of peers via beacon-to-app communication and the participant's EMA report of peer presence. Disagreement between the beacon and self-report data (ie, false negatives and false positives) will be investigated in multiple ways (ie, to determine if the reason was technology-related or participant compliance-related) using encounter data and information collected from participants and peers. RESULTS: Participant recruitment began in February 2023, and enrollment was completed in December 2023. Results will be reported in 2025. CONCLUSIONS: This Bluetooth-based technology has important applications and clinical implications for various health behaviors, including the potential for just-in-time adaptive interventions that target high-risk behavior in real time. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/50241.

Personalized AI-Driven Real-Time Models to Predict Stress-Induced Blood Pressure Spikes Using Wearable Devices: Proposal for a Prospective Cohort Study.

BACKGROUND: Referred to as the "silent killer," elevated blood pressure (BP) often goes unnoticed due to the absence of apparent symptoms, resulting in cumulative harm over time. Chronic stress has been consistently linked to increased BP. Prior studies have found that elevated BP often arises due to a stressful lifestyle, although the effect of exact stressors varies drastically between individuals. The heterogeneous nature of both the stress and BP response to a multitude of lifestyle decisions can make it difficult if not impossible to pinpoint the most deleterious behaviors using the traditional mechanism of clinical interviews. OBJECTIVE: The aim of this study is to leverage machine learning (ML) algorithms for real-time predictions of stress-induced BP spikes using consumer wearable devices such as Fitbit, providing actionable insights to both patients and clinicians to improve diagnostics and enable proactive health monitoring. This study also seeks to address the significant challenges in identifying specific deleterious behaviors associated with stress-induced hypertension through the development of personalized artificial intelligence models for individual patients, departing from the conventional approach of using generalized models. METHODS: The study proposes the development of ML algorithms to analyze biosignals obtained from these wearable devices, aiming to make real-time predictions about BP spikes. Given the longitudinal nature of the data set comprising time-series data from wearables (eg, Fitbit) and corresponding time-stamped labels representing stress levels from Ecological Momentary Assessment reports, the adoption of self-supervised learning for pretraining the network and using transformer models for fine-tuning the model on a personalized prediction task is proposed. Transformer models, with their self-attention mechanisms, dynamically weigh the importance of different time steps, enabling the model to focus on relevant temporal features and dependencies, facilitating accurate prediction. RESULTS: Supported as a pilot project from the Robert C Perry Fund of the Hawaii Community Foundation, the study team has developed the core study app, CardioMate. CardioMate not only reminds participants to initiate BP readings using an Omron HeartGuide wearable monitor but also prompts them multiple times a day to report stress levels. Additionally, it collects other useful information including medications, environmental conditions, and daily interactions. Through the app's messaging system, efficient contact and interaction between users and study admins ensure smooth progress. CONCLUSIONS: Personalized ML when applied to biosignals offers the potential for real-time digital health interventions for chronic stress and its symptoms. The project's clinical use for Hawaiians with stress-induced high BP combined with its methodological innovation of personalized artificial intelligence models highlights its significance in advancing health care interventions. Through iterative refinement and optimization, the aim is to develop a personalized deep-learning framework capable of accurately predicting stress-induced BP spikes, thereby promoting individual well-being and health outcomes. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/55615.

Personalized Deep Learning for Substance Use in Hawaii: Protocol for a Passive Sensing and Ecological Momentary Assessment Study.

BACKGROUND: Artificial intelligence (AI)-powered digital therapies that detect methamphetamine cravings via consumer devices have the potential to reduce health care disparities by providing remote and accessible care solutions to communities with limited care solutions, such as Native Hawaiian, Filipino, and Pacific Islander communities. However, Native Hawaiian, Filipino, and Pacific Islander communities are understudied with respect to digital therapeutics and AI health sensing despite using technology at the same rates as other racial groups. OBJECTIVE: In this study, we aimed to understand the feasibility of continuous remote digital monitoring and ecological momentary assessments in Native Hawaiian, Filipino, and Pacific Islander communities in Hawaii by curating a novel data set of longitudinal Fitbit (Fitbit Inc) biosignals with the corresponding craving and substance use labels. We also aimed to develop personalized AI models that predict methamphetamine craving events in real time using wearable sensor data. METHODS: We will develop personalized AI and machine learning models for methamphetamine use and craving prediction in 40 individuals from Native Hawaiian, Filipino, and Pacific Islander communities by curating a novel data set of real-time Fitbit biosensor readings and the corresponding participant annotations (ie, raw self-reported substance use data) of their methamphetamine use and cravings. In the process of collecting this data set, we will gain insights into cultural and other human factors that can challenge the proper acquisition of precise annotations. With the resulting data set, we will use self-supervised learning AI approaches, which are a new family of machine learning methods that allows a neural network to be trained without labels by being optimized to make predictions about the data. The inputs to the proposed AI models are Fitbit biosensor readings, and the outputs are predictions of methamphetamine use or craving. This paradigm is gaining increased attention in AI for health care. RESULTS: To date, more than 40 individuals have expressed interest in participating in the study, and we have successfully recruited our first 5 participants with minimal logistical challenges and proper compliance. Several logistical challenges that the research team has encountered so far and the related implications are discussed. CONCLUSIONS: We expect to develop models that significantly outperform traditional supervised methods by finetuning according to the data of a participant. Such methods will enable AI solutions that work with the limited data available from Native Hawaiian, Filipino, and Pacific Islander populations and that are inherently unbiased owing to their personalized nature. Such models can support future AI-powered digital therapeutics for substance abuse. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/46493.

Machine Learning for Multimodal Mental Health Detection: A Systematic Review of Passive Sensing Approaches.

As mental health (MH) disorders become increasingly prevalent, their multifaceted symptoms and comorbidities with other conditions introduce complexity to diagnosis, posing a risk of underdiagnosis. While machine learning (ML) has been explored to mitigate these challenges, we hypothesized that multiple data modalities support more comprehensive detection and that non-intrusive collection approaches better capture natural behaviors. To understand the current trends, we systematically reviewed 184 studies to assess feature extraction, feature fusion, and ML methodologies applied to detect MH disorders from passively sensed multimodal data, including audio and video recordings, social media, smartphones, and wearable devices. Our findings revealed varying correlations of modality-specific features in individualized contexts, potentially influenced by demographics and personalities. We also observed the growing adoption of neural network architectures for model-level fusion and as ML algorithms, which have demonstrated promising efficacy in handling high-dimensional features while modeling within and cross-modality relationships. This work provides future researchers with a clear taxonomy of methodological approaches to multimodal detection of MH disorders to inspire future methodological advancements. The comprehensive analysis also guides and supports future researchers in making informed decisions to select an optimal data source that aligns with specific use cases based on the MH disorder of interest.

Health Care Professionals' Views on the Use of Passive Sensing, AI, and Machine Learning in Mental Health Care: Systematic Review With Meta-Synthesis.

BACKGROUND: Mental health difficulties are highly prevalent worldwide. Passive sensing technologies and applied artificial intelligence (AI) methods can provide an innovative means of supporting the management of mental health problems and enhancing the quality of care. However, the views of stakeholders are important in understanding the potential barriers to and facilitators of their implementation. OBJECTIVE: This study aims to review, critically appraise, and synthesize qualitative findings relating to the views of mental health care professionals on the use of passive sensing and AI in mental health care. METHODS: A systematic search of qualitative studies was performed using 4 databases. A meta-synthesis approach was used, whereby studies were analyzed using an inductive thematic analysis approach within a critical realist epistemological framework. RESULTS: Overall, 10 studies met the eligibility criteria. The 3 main themes were uses of passive sensing and AI in clinical practice, barriers to and facilitators of use in practice, and consequences for service users. A total of 5 subthemes were identified: barriers, facilitators, empowerment, risk to well-being, and data privacy and protection issues. CONCLUSIONS: Although clinicians are open-minded about the use of passive sensing and AI in mental health care, important factors to consider are service user well-being, clinician workloads, and therapeutic relationships. Service users and clinicians must be involved in the development of digital technologies and systems to ensure ease of use. The development of, and training in, clear policies and guidelines on the use of passive sensing and AI in mental health care, including risk management and data security procedures, will also be key to facilitating clinician engagement. The means for clinicians and service users to provide feedback on how the use of passive sensing and AI in practice is being received should also be considered. TRIAL REGISTRATION: PROSPERO International Prospective Register of Systematic Reviews CRD42022331698; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=331698.

Soft Robotic Finger with Energy-Coupled Quadrastability.

The performance of the human finger is a significant inspiration for designing soft robotic fingers that can achieve high speed and high force or perform delicate and complex tasks. Existing soft grippers and actuators can be excellent in specific capabilities. However, it is still challenging for them to meet an all-around performance as the human finger, characterized by high actuation speed, wide grasping range, sensing ability, and gentle and high-load grasping capability. The proposed tendon pulley quadrastable (TPQ) finger has combined these qualities in the conducted gripping tasks. A pair of elastic tendons is utilized as the sole energy reservoir to create a novel energy distribution pattern: energy-coupled quadrastability. An energy model is built to analyze and predict the behaviors of the TPQ finger. Mechanical instability is utilized to enhance the actuation speed. The proposed soft lever mechanism endows the TPQ finger with sensing ability. The energy barrier adjusting plates control the energy barrier, adjusting the sensitivity of both active and passive actuation mechanisms. The transition of four stable states forms preplanned trajectories that are applied to create multiple grasping manners. Experiments show that it can respond to stimuli and finish a grasping task in merely 31 ms, and its payload can reach 33.25 kg. At the same time, it can also handle fragile objects such as a piece of rose and grasp a wide range of objects ranging from a thin nut (3.3 mm in height) or a thin card (0.76 mm thick) to a football (220 mm).

Digital Phenotyping for Monitoring Mental Disorders: Systematic Review.

BACKGROUND: The COVID-19 pandemic has increased the impact and spread of mental illness and made health services difficult to access; therefore, there is a need for remote, pervasive forms of mental health monitoring. Digital phenotyping is a new approach that uses measures extracted from spontaneous interactions with smartphones (eg, screen touches or movements) or other digital devices as markers of mental status. OBJECTIVE: This review aimed to evaluate the feasibility of using digital phenotyping for predicting relapse or exacerbation of symptoms in patients with mental disorders through a systematic review of the scientific literature. METHODS: Our research was carried out using 2 bibliographic databases (PubMed and Scopus) by searching articles published up to January 2023. By following the PRISMA (Preferred Reporting Items for Systematic Review and Meta-Analysis) guidelines, we started from an initial pool of 1150 scientific papers and screened and extracted a final sample of 29 papers, including studies concerning clinical populations in the field of mental health, which were aimed at predicting relapse or exacerbation of symptoms. The systematic review has been registered on the web registry Open Science Framework. RESULTS: We divided the results into 4 groups according to mental disorder: schizophrenia (9/29, 31%), mood disorders (15/29, 52%), anxiety disorders (4/29, 14%), and substance use disorder (1/29, 3%). The results for the first 3 groups showed that several features (ie, mobility, location, phone use, call log, heart rate, sleep, head movements, facial and vocal characteristics, sociability, social rhythms, conversations, number of steps, screen on or screen off status, SMS text message logs, peripheral skin temperature, electrodermal activity, light exposure, and physical activity), extracted from data collected via the smartphone and wearable wristbands, can be used to create digital phenotypes that could support gold-standard assessment and could be used to predict relapse or symptom exacerbations. CONCLUSIONS: Thus, as the data were consistent for almost all the mental disorders considered (mood disorders, anxiety disorders, and schizophrenia), the feasibility of this approach was confirmed. In the future, a new model of health care management using digital devices should be integrated with the digital phenotyping approach and tailored mobile interventions (managing crises during relapse or exacerbation).

A Passive RF Testbed for Human Posture Classification in FM Radio Bands.

This paper explores the opportunities and challenges for classifying human posture in indoor scenarios by analyzing the Frequency-Modulated (FM) radio broadcasting signal received at multiple locations. More specifically, we present a passive RF testbed operating in FM radio bands, which allows experimentation with innovative human posture classification techniques. After introducing the details of the proposed testbed, we describe a simple methodology to detect and classify human posture. The methodology includes a detailed study of feature engineering and the assumption of three traditional classification techniques. The implementation of the proposed methodology in software-defined radio devices allows an evaluation of the testbed's capability to classify human posture in real time. The evaluation results presented in this paper confirm that the accuracy of the classification can be approximately 90%, showing the effectiveness of the proposed testbed and its potential to support the development of future innovative classification techniques by only sensing FM bands in a passive mode.

The relationship between loneliness and depression among college students: Mining data derived from passive sensing.

Background: While there is recognition of the relationship between loneliness and depression, specific behavioural patterns distinguishing both are still not well understood. Objectives: Our objective is to identify distinct behavioural patterns of loneliness and depression from a passively collected dataset of college students, understand their similarities and interrelationships and assess their effectiveness in identifying loneliness and depression. Methods: Utilizing the StudentLife dataset, we applied regression analysis to determine associations with self-reported loneliness and depression. Mediation analysis interprets the relationship between the two conditions, and machine learning models predict loneliness and depression based on behavioural indicators. Results: Distinct behavioural patterns emerged: high evening screen time (OR = 1.45, p = 0.002) and high overall phone usage (OR = 1.50, p = 0.003) were associated with more loneliness, whereas depression was significantly associated with fewer screen unlocks (OR = 0.75, p = 0.044) and visits to fewer unique places (OR = 0.85, p = 0.023). Longer durations of physical activity (OR = 0.72, p = 0.014) and sleep (OR = 0.46, p = 0.002) are linked to a lower risk of both loneliness and depression. Mediation analysis revealed that loneliness significantly increases the likelihood of depression by 48%. The prediction accuracy of our XGBoost-based machine learning approach was 82.43% for loneliness and 79.43% for depression. Conclusion: Our findings show that high evening screen time and overall phone usage are significantly associated with increased loneliness, while fewer screen unlocks and visits to fewer unique places are significantly related to depression. The findings can help in developing targeted interventions to promote well-being and mental health in students.