Passive Sensing of Prediction of Moment-To-Moment Depressed Mood among Undergraduates with Clinical Levels of Depression Sample Using Smartphones

Predicting Mood Based on the Social Context Measured Through the Experience Sampling Method, Digital Phenotyping, and Social Networks

Social interactions are essential for well-being. Therefore, researchers increasingly attempt to capture an individual's social context to predict well-being, including mood. Different tools are used to measure various aspects of the social context. Digital phenotyping is a commonly used technology to assess a person's social behavior objectively. The experience sampling method (ESM) can capture the subjective perception of specific interactions. Lastly, egocentric networks are often used to measure specific relationship characteristics. These different methods capture different aspects of the social context over different time scales that are related to well-being, and combining them may be necessary to improve the prediction of well-being. Yet, they have rarely been combined in previous research. To address this gap, our study investigates the predictive accuracy of mood based on the social context. We collected intensive within-person data from multiple passive and self-report sources over a 28-day period in a student sample (Participants: N = 11, ESM measures: N = 1313). We trained individualized random forest machine learning models, using different predictors included in each model summarized over different time scales. Our findings revealed that even when combining social interactions data using different methods, predictive accuracy of mood remained low. The average coefficient of determination over all participants was 0.06 for positive and negative affect and ranged from - 0.08 to 0.3, indicating a large amount of variance across people. Furthermore, the optimal set of predictors varied across participants; however, predicting mood using all predictors generally yielded the best predictions. While combining different predictors improved predictive accuracy of mood for most participants, our study highlights the need for further work using larger and more diverse samples to enhance the clinical utility of these predictive modeling approaches.

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.

Effectiveness of artificial intelligence in detecting and managing depressive disorders: Systematic review

OBJECTIVES

This study underscores the importance of exploring AI's creative applications in treating depressive disorders to revolutionize mental health care. Through innovative integration of AI technologies, the research confirms their positive effects on preventing, diagnosing, and treating depression. The systematic review establishes an evidence base for AI in depression management, offering directions for effective interventions.

METHODS

This systematic literature review investigates the effectiveness of AI in depression management by analyzing studies from January 1, 2017, to May 31, 2022. Utilizing search engines like IEEE Xplore, PubMed, and Web of Science, the review focused on keywords such as Depression/Mental Health, Machine Learning/Artificial Intelligence, and Prediction/Diagnosis. The analysis of 95 documents involved classification based on use, data type, and algorithm type.

RESULTS

The study revealed that AI in depression management excelled in accuracy, particularly in monitoring and prediction. Biomarker-derived data demonstrated the highest accuracy, with the CNN algorithm proving most effective. The findings affirm the therapeutic benefits of AI, including treatment, detection, and disease prediction, highlighting its potential in analyzing monitored data for depression management.

LIMITATIONS

This study exclusively examined the application of AI in individuals with depressive disorders. Interpretation should be cautious due to the limited scope of subjects to this specific population.

CONCLUSIONS

To introduce digital healthcare and therapies for ongoing depression management, it's crucial to present empirical evidence on the medical fee payment system, safety, and efficacy. These findings support enhanced medical accessibility through digital healthcare, offering personalized disease management for patients seeking non-face-to-face treatment.

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.

Personalized Machine Learning-Based Prediction of Wellbeing and Empathy in Healthcare Professionals

Healthcare professionals are known to suffer from workplace stress and burnout, which can negatively affect their empathy for patients and quality of care. While existing research has identified factors associated with wellbeing and empathy in healthcare professionals, these efforts are typically focused on the group level, ignoring potentially important individual differences and implications for individualized intervention approaches. In the current study, we implemented N-of-1 personalized machine learning (PML) to predict wellbeing and empathy in healthcare professionals at the individual level, leveraging ecological momentary assessments (EMAs) and smartwatch wearable data. A total of 47 mood and lifestyle feature variables (relating to sleep, diet, exercise, and social connections) were collected daily for up to three months followed by applying eight supervised machine learning (ML) models in a PML pipeline to predict wellbeing and empathy separately. Predictive insight into the model architecture was obtained using Shapley statistics for each of the best-fit personalized models, ranking the importance of each feature for each participant. The best-fit model and top features varied across participants, with anxious mood (13/19) and depressed mood (10/19) being the top predictors in most models. Social connection was a top predictor for wellbeing in 9/12 participants but not for empathy models (1/7). Additionally, empathy and wellbeing were the top predictors of each other in 64% of cases. These findings highlight shared and individual features of wellbeing and empathy in healthcare professionals and suggest that a one-size-fits-all approach to addressing modifiable factors to improve wellbeing and empathy will likely be suboptimal. In the future, such personalized models may serve as actionable insights for healthcare professionals that lead to increased wellness and quality of patient care.

Benchmarking Mental Health Status Using Passive Sensor Data: Protocol for a Prospective Observational Study

BACKGROUND

Computational psychiatry has the potential to advance the diagnosis, mechanistic understanding, and treatment of mental health conditions. Promising results from clinical samples have led to calls to extend these methods to mental health risk assessment in the general public; however, data typically used with clinical samples are neither available nor scalable for research in the general population. Digital phenotyping addresses this by capitalizing on the multimodal and widely available data created by sensors embedded in personal digital devices (eg, smartphones) and is a promising approach to extending computational psychiatry methods to improve mental health risk assessment in the general population.

OBJECTIVE

Building on recommendations from existing computational psychiatry and digital phenotyping work, we aim to create the first computational psychiatry data set that is tailored to studying mental health risk in the general population; includes multimodal, sensor-based behavioral features; and is designed to be widely shared across academia, industry, and government using gold standard methods for privacy, confidentiality, and data integrity.

METHODS

We are using a stratified, random sampling design with 2 crossed factors (difficulties with emotion regulation and perceived life stress) to recruit a sample of 400 community-dwelling adults balanced across high- and low-risk for episodic mental health conditions. Participants first complete self-report questionnaires assessing current and lifetime psychiatric and medical diagnoses and treatment, and current psychosocial functioning. Participants then complete a 7-day in situ data collection phase that includes providing daily audio recordings, passive sensor data collected from smartphones, self-reports of daily mood and significant events, and a verbal description of the significant daily events during a nightly phone call. Participants complete the same baseline questionnaires 6 and 12 months after this phase. Self-report questionnaires will be scored using standard methods. Raw audio and passive sensor data will be processed to create a suite of daily summary features (eg, time spent at home).

RESULTS

Data collection began in June 2022 and is expected to conclude by July 2024. To date, 310 participants have consented to the study; 149 have completed the baseline questionnaire and 7-day intensive data collection phase; and 61 and 31 have completed the 6- and 12-month follow-up questionnaires, respectively. Once completed, the proposed data set will be made available to academic researchers, industry, and the government using a stepped approach to maximize data privacy.

CONCLUSIONS

This data set is designed as a complementary approach to current computational psychiatry and digital phenotyping research, with the goal of advancing mental health risk assessment within the general population. This data set aims to support the field's move away from siloed research laboratories collecting proprietary data and toward interdisciplinary collaborations that incorporate clinical, technical, and quantitative expertise at all stages of the research process.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/53857.

Predicting daily cognition and lifestyle behaviors for older adults using smart home data and ecological momentary assessment

OBJECTIVE

Extraction of digital markers from passive sensors placed in homes is a promising method for understanding real-world behaviors. In this study, machine learning (ML) and multilevel modeling (MLM) are used to examine types of digital markers and whether smart home sensors can predict cognitive functioning, lifestyle behaviors, and contextual factors measured through ecological momentary assessment (EMA).

METHOD

Smart home sensors were installed in the homes of 44 community-dwelling midlife and older adults for 3-4 months. Sensor data were categorized into eight digital markers. Participants responded to iPad-delivered EMA prompts 4×/day for 2 wk. Prompts included an n-back task and survey on recent (past 2 h) lifestyle and contextual factors.

RESULTS

ML marker rankings revealed that sensor counts (indicating increased activity) and time outside the home were among the most influential markers for all survey questions. Additionally, MLM revealed for every 1000 sensor counts, mental sharpness, social, physical, and cognitive EMA responses increased by 0.134-0.155 points on a 5-point scale. For every additional 30-minutes spent outside home, social, physical, and cognitive EMA responses increased by 0.596, 0.472, and 0.157 points. Advanced ML joint classification/regression significantly predicted EMA responses from smart home digital markers with error of 0.370 on a 5-point scale, and n-back performance with a normalized error of 0.040.

CONCLUSION

Results from ML and MLM were complimentary and comparable, suggesting that machine learning may be used to develop generalized models to predict everyday cognition and track lifestyle behaviors and contextual factors that impact health outcomes using smart home sensor data.

A collaborative realist review of remote measurement technologies for depression in young people

Digital mental health is becoming increasingly common. This includes use of smartphones and wearables to collect data in real time during day-to-day life (remote measurement technologies, RMT). Such data could capture changes relevant to depression for use in objective screening, symptom management and relapse prevention. This approach may be particularly accessible to young people of today as the smartphone generation. However, there is limited research on how such a complex intervention would work in the real world. We conducted a collaborative realist review of RMT for depression in young people. Here we describe how, why, for whom and in what contexts RMT appear to work or not work for depression in young people and make recommendations for future research and practice. Ethical, data protection and methodological issues need to be resolved and standardized; without this, RMT may be currently best used for self-monitoring and feedback to the healthcare professional where possible, to increase emotional self-awareness, enhance the therapeutic relationship and monitor the effectiveness of other interventions.

Differential temporal utility of passively sensed smartphone features for depression and anxiety symptom prediction: a longitudinal cohort study

While studies show links between smartphone data and affective symptoms, we lack clarity on the temporal scale, specificity (e.g., to depression vs. anxiety), and person-specific (vs. group-level) nature of these associations. We conducted a large-scale (n = 1013) smartphone-based passive sensing study to identify within- and between-person digital markers of depression and anxiety symptoms over time. Participants (74.6% female; M age = 40.9) downloaded the LifeSense app, which facilitated continuous passive data collection (e.g., GPS, app and device use, communication) across 16 weeks. Hierarchical linear regression models tested the within- and between-person associations of 2-week windows of passively sensed data with depression (PHQ-8) or generalized anxiety (GAD-7). We used a shifting window to understand the time scale at which sensed features relate to mental health symptoms, predicting symptoms 2 weeks in the future (distal prediction), 1 week in the future (medial prediction), and 0 weeks in the future (proximal prediction). Spending more time at home relative to one's average was an early signal of PHQ-8 severity (distal β = 0.219, p = 0.012) and continued to relate to PHQ-8 at medial (β = 0.198, p = 0.022) and proximal (β = 0.183, p = 0.045) windows. In contrast, circadian movement was proximally related to (β = -0.131, p = 0.035) but did not predict (distal β = 0.034, p = 0.577; medial β = -0.089, p = 0.138) PHQ-8. Distinct communication features (i.e., call/text or app-based messaging) related to PHQ-8 and GAD-7. Findings have implications for identifying novel treatment targets, personalizing digital mental health interventions, and enhancing traditional patient-provider interactions. Certain features (e.g., circadian movement) may represent correlates but not true prospective indicators of affective symptoms. Conversely, other features like home duration may be such early signals of intra-individual symptom change, indicating the potential utility of prophylactic intervention (e.g., behavioral activation) in response to person-specific increases in these signals.

Towards Personalised Mood Prediction and Explanation for Depression from Biophysical Data

Digital health applications using Artificial Intelligence (AI) are a promising opportunity to address the widening gap between available resources and mental health needs globally. Increasingly, passively acquired data from wearables are augmented with carefully selected active data from depressed individuals to develop Machine Learning (ML) models of depression based on mood scores. However, most ML models are black box in nature, and hence the outputs are not explainable. Depression is also multimodal, and the reasons for depression may vary significantly between individuals. Explainable and personalised models will thus be beneficial to clinicians to determine the main features that lead to a decline in the mood state of a depressed individual, thus enabling suitable personalised therapy. This is currently lacking. Therefore, this study presents a methodology for developing personalised and accurate Deep Learning (DL)-based predictive mood models for depression, along with novel methods for identifying the key facets that lead to the exacerbation of depressive symptoms. We illustrate our approach by using an existing multimodal dataset containing longitudinal Ecological Momentary Assessments of depression, lifestyle data from wearables and neurocognitive assessments for 14 mild to moderately depressed participants over one month. We develop classification- and regression-based DL models to predict participants' current mood scores-a discrete score given to a participant based on the severity of their depressive symptoms. The models are trained inside eight different evolutionary-algorithm-based optimisation schemes that optimise the model parameters for a maximum predictive performance. A five-fold cross-validation scheme is used to verify the DL model's predictive performance against 10 classical ML-based models, with a model error as low as 6% for some participants. We use the best model from the optimisation process to extract indicators, using SHAP, ALE and Anchors from explainable AI literature to explain why certain predictions are made and how they affect mood. These feature insights can assist health professionals in incorporating personalised interventions into a depressed individual's treatment regimen.

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 smartphone- and wearable-based biomarker for the estimation of unipolar depression severity

Drug development for mood disorders can greatly benefit from the development of robust, reliable, and objective biomarkers. The incorporation of smartphones and wearable devices in clinical trials provide a unique opportunity to monitor behavior in a non-invasive manner. The objective of this study is to identify the correlations between remotely monitored self-reported assessments and objectively measured activities with depression severity assessments often applied in clinical trials. 30 unipolar depressed patients and 29 age- and gender-matched healthy controls were enrolled in this study. Each participant's daily physiological, physical, and social activity were monitored using a smartphone-based application (CHDR MORE™) for 3 weeks continuously. Self-reported depression anxiety stress scale-21 (DASS-21) and positive and negative affect schedule (PANAS) were administered via smartphone weekly and daily respectively. The structured interview guide for the Hamilton depression scale and inventory of depressive symptomatology-clinical rated (SIGHD-IDSC) was administered in-clinic weekly. Nested cross-validated linear mixed-effects models were used to identify the correlation between the CHDR MORE™ features with the weekly in-clinic SIGHD-IDSC scores. The SIGHD-IDSC regression model demonstrated an explained variance (R2) of 0.80, and a Root Mean Square Error (RMSE) of ± 15 points. The SIGHD-IDSC total scores were positively correlated with the DASS and mean steps-per-minute, and negatively correlated with the travel duration. Unobtrusive, remotely monitored behavior and self-reported outcomes are correlated with depression severity. While these features cannot replace the SIGHD-IDSC for estimating depression severity, it can serve as a complementary approach for assessing depression and drug effects outside the clinic.

Improving Depression Severity Prediction from Passive Sensing: Symptom-Profiling Approach

Depression is a significant mental health issue that profoundly impacts people's lives. Diagnosing depression often involves interviews with mental health professionals and surveys, which can become cumbersome when administered continuously. Digital phenotyping offers an innovative approach for detecting and monitoring depression without requiring active user involvement. This study contributes to the detection of depression severity and depressive symptoms using mobile devices. Our proposed approach aims to distinguish between different patterns of depression and improve prediction accuracy. We conducted an experiment involving 381 participants over a period of at least three months, during which we collected comprehensive passive sensor data and Patient Health Questionnaire (PHQ-9) self-reports. To enhance the accuracy of predicting depression severity levels (classified as none/mild, moderate, or severe), we introduce a novel approach called symptom profiling. The symptom profile vector represents nine depressive symptoms and indicates both the probability of each symptom being present and its significance for an individual. We evaluated the effectiveness of the symptom-profiling method by comparing the F1 score achieved using sensor data features as inputs to machine learning models with the F1 score obtained using the symptom profile vectors as inputs. Our findings demonstrate that symptom profiling improves the F1 score by up to 0.09, with an average improvement of 0.05, resulting in a depression severity prediction with an F1 score as high as 0.86.

Ensemble Approach to Combining Episode Prediction Models Using Sequential Circadian Rhythm Sensor Data from Mental Health Patients

Managing mood disorders poses challenges in counseling and drug treatment, owing to limitations. Counseling is the most effective during hospital visits, and the side effects of drugs can be burdensome. Patient empowerment is crucial for understanding and managing these triggers. The daily monitoring of mental health and the utilization of episode prediction tools can enable self-management and provide doctors with insights into worsening lifestyle patterns. In this study, we test and validate whether the prediction of future depressive episodes in individuals with depression can be achieved by using lifelog sequence data collected from digital device sensors. Diverse models such as random forest, hidden Markov model, and recurrent neural network were used to analyze the time-series data and make predictions about the occurrence of depressive episodes in the near future. The models were then combined into a hybrid model. The prediction accuracy of the hybrid model was 0.78; especially in the prediction of rare episode events, the F1-score performance was approximately 1.88 times higher than that of the dummy model. We explored factors such as data sequence size, train-to-test data ratio, and class-labeling time slots that can affect the model performance to determine the combinations of parameters that optimize the model performance. Our findings are especially valuable because they are experimental results derived from large-scale participant data analyzed over a long period of time.

Predicting Individual Response to a Web-Based Positive Psychology Intervention: A Machine Learning Approach

Positive psychology interventions (PPIs) are effective at increasing happiness and decreasing depressive symptoms. PPIs are often administered as self-guided web-based interventions, but not all persons benefit from web-based interventions. Therefore, it is important to identify whether someone is likely to benefit from web-based PPIs, in order to triage persons who may not benefit from other interventions. In the current study, we used machine learning to predict individual response to a web-based PPI, in order to investigate baseline prognostic indicators of likelihood of response (N = 120). Our models demonstrated moderate correlations (happiness: r Test = 0.30 ± 0.09; depressive symptoms: r Test = 0.39 ± 0.06), indicating that baseline features can predict changes in happiness and depressive symptoms at a 6-month follow-up. Thus, machine learning can be used to predict outcome changes from a web-based PPI and has important clinical implications for matching individuals to PPIs based on their individual characteristics.

Depression deconstructed: Wearables and passive digital phenotyping for analyzing individual symptoms

Wearable technology enables unobtrusive collection of longitudinally dense data, allowing for continuous monitoring of physiology and behavior. These digital phenotypes, or device-based indicators, are frequently leveraged to study depression. However, they are usually considered alongside questionnaire sum-scores which collapse the symptomatic gamut into a general representation of severity. To explore the contributions of passive sensing streams more precisely, associations of nine passive sensing-derived features with self-report responses to Center for Epidemiologic Studies Depression (CES-D) items were modeled. Using data from the NetHealth study on N=469 college students, this work generated mixed ordinal logistic regression models to summarize contributions of pulse, movement, and sleep data to depression symptom detection. Emphasizing the importance of the college context, wearable features displayed unique and complementary properties in their heterogeneously significant associations with CES-D items. This work provides conceptual and exploratory blueprints for a reductionist approach to modeling depression within passive sensing research.

Acceptability of Personal Sensing Among People With Alcohol Use Disorder: Observational Study

BACKGROUND

Personal sensing may improve digital therapeutics for mental health care by facilitating early screening, symptom monitoring, risk prediction, and personalized adaptive interventions. However, further development and the use of personal sensing requires a better understanding of its acceptability to people targeted for these applications.

OBJECTIVE

We aimed to assess the acceptability of active and passive personal sensing methods in a sample of people with moderate to severe alcohol use disorder using both behavioral and self-report measures. This sample was recruited as part of a larger grant-funded project to develop a machine learning algorithm to predict lapses.

METHODS

Participants (N=154; n=77, 50% female; mean age 41, SD 11.9 years; n=134, 87% White and n=150, 97% non-Hispanic) in early recovery (1-8 weeks of abstinence) were recruited to participate in a 3-month longitudinal study. Participants were modestly compensated for engaging with active (eg, ecological momentary assessment [EMA], audio check-in, and sleep quality) and passive (eg, geolocation, cellular communication logs, and SMS text message content) sensing methods that were selected to tap into constructs from the Relapse Prevention model by Marlatt. We assessed 3 behavioral indicators of acceptability: participants' choices about their participation in the study at various stages in the procedure, their choice to opt in to provide data for each sensing method, and their adherence to a subset of the active methods (EMA and audio check-in). We also assessed 3 self-report measures of acceptability (interference, dislike, and willingness to use for 1 year) for each method.

RESULTS

Of the 192 eligible individuals screened, 191 consented to personal sensing. Most of these individuals (169/191, 88.5%) also returned 1 week later to formally enroll, and 154 participated through the first month follow-up visit. All participants in our analysis sample opted in to provide data for EMA, sleep quality, geolocation, and cellular communication logs. Out of 154 participants, 1 (0.6%) did not provide SMS text message content and 3 (1.9%) did not provide any audio check-ins. The average adherence rate for the 4 times daily EMA was .80. The adherence rate for the daily audio check-in was .54. Aggregate participant ratings indicated that all personal sensing methods were significantly more acceptable (all P<.001) compared with neutral across subjective measures of interference, dislike, and willingness to use for 1 year. Participants did not significantly differ in their dislike of active methods compared with passive methods (P=.23). However, participants reported a higher willingness to use passive (vs active) methods for 1 year (P=.04).

CONCLUSIONS

These results suggest that active and passive sensing methods are acceptable for people with alcohol use disorder over a longer period than has previously been assessed. Important individual differences were observed across people and methods, indicating opportunities for future improvement.

Predicting states of elevated negative affect in adolescents from smartphone sensors: a novel personalized machine learning approach

BACKGROUND

Adolescence is characterized by profound change, including increases in negative emotions. Approximately 84% of American adolescents own a smartphone, which can continuously and unobtrusively track variables potentially predictive of heightened negative emotions (e.g. activity levels, location, pattern of phone usage). The extent to which built-in smartphone sensors can reliably predict states of elevated negative affect in adolescents is an open question.

METHODS

Adolescent participants (n = 22; ages 13-18) with low to high levels of depressive symptoms were followed for 15 weeks using a combination of ecological momentary assessments (EMAs) and continuously collected passive smartphone sensor data. EMAs probed negative emotional states (i.e. anger, sadness and anxiety) 2-3 times per day every other week throughout the study (total: 1145 EMA measurements). Smartphone accelerometer, location and device state data were collected to derive 14 discrete estimates of behavior, including activity level, percentage of time spent at home, sleep onset and duration, and phone usage.

RESULTS

A personalized ensemble machine learning model derived from smartphone sensor data outperformed other statistical approaches (e.g. linear mixed model) and predicted states of elevated anger and anxiety with acceptable discrimination ability (area under the curve (AUC) = 74% and 71%, respectively), but demonstrated more modest discrimination ability for predicting states of high sadness (AUC = 66%).

CONCLUSIONS

To the extent that smartphone data could provide reasonably accurate real-time predictions of states of high negative affect in teens, brief 'just-in-time' interventions could be immediately deployed via smartphone notifications or mental health apps to alleviate these states.

A Systematic Review of Location Data for Depression Prediction

Depression contributes to a wide range of maladjustment problems. With the development of technology, objective measurement for behavior and functional indicators of depression has become possible through the passive sensing technology of digital devices. Focusing on location data, we systematically reviewed the relationship between depression and location data. We searched Scopus, PubMed, and Web of Science databases by combining terms related to passive sensing and location data with depression. Thirty-one studies were included in this review. Location data demonstrated promising predictive power for depression. Studies examining the relationship between individual location data variables and depression, homestay, entropy, and the normalized entropy variable of entropy dimension showed the most consistent and significant correlations. Furthermore, variables of distance, irregularity, and location showed significant associations in some studies. However, semantic location showed inconsistent results. This suggests that the process of geographical movement is more related to mood changes than to semantic location. Future research must converge across studies on location-data measurement methods.

Understanding the construction of 'behavior' in smartphone addiction: A scoping review

It has been claimed that smartphone usage constitutes a behavioral addiction, characterised by compulsive, excessive use of one's phone and psychological withdrawal or distress when the phone is absent. However, there is uncertainty about key phenomenological and conceptual details of smartphone addiction. One of the central problems has been understanding the processes that link smartphone usage, and addiction. The question this paper aims to answer is straightforward: based on measures utilised in the literature, what does 'behavior' mean in the context of smartphone addiction? A scoping review of the smartphone addiction literature was undertaken. This identified 1305 studies collecting smartphone addiction data. Just under half (49.89%) of all published smartphone addiction papers did not report the collection of any smartphone specific behaviors. Those that did tended to focus on a small cluster of self-reported behaviors capturing volume of overall use: hours spent using a smartphone per day, number of pickups, duration of smartphone ownership, and types of app used. Approximately 10% of papers used logged behavioral data on phones. Although the theoretical literature places increasing focus on context and patterns of use, measurements of behavior tend to focus on broad, volumetric measures. The number of studies reporting behavior has decreased over time, suggesting smartphone addiction is becoming increasingly trait-like. Both major phone operating systems have proprietary apps that collected behavioral data by default, and research in the field should take advantage of these capabilities when measuring smartphone usage.

Digital Phenotyping of Mental Health using multimodal sensing of multiple situations of interest: A Systematic Literature Review

Many studies have used Digital Phenotyping of Mental Health (DPMH) to complement classic methods of mental health assessment and monitoring. This research area proposes innovative methods that perform multimodal sensing of multiple situations of interest (e.g., sleep, physical activity, mobility) to health professionals. In this paper, we present a Systematic Literature Review (SLR) to recognize, characterize and analyze the state of the art on DPMH using multimodal sensing of multiple situations of interest to professionals. We searched for studies in six digital libraries, which resulted in 1865 retrieved published papers. Next, we performed a systematic process of selecting studies based on inclusion and exclusion criteria, which selected 59 studies for the data extraction phase. First, based on the analysis of the extracted data, we describe an overview of this field, then presenting characteristics of the selected studies, the main mental health topics targeted, the physical and virtual sensors used, and the identified situations of interest. Next, we outline answers to research questions, describing the context data sources used to detect situations, the DPMH workflow used for multimodal sensing of situations, and the application of DPMH solutions in the mental health assessment and monitoring process. In addition, we recognize trends presented by DPMH studies, such as the design of solutions for high-level information recognition, association of features with mental states/disorders, classification of mental states/disorders, and prediction of mental states/disorders. We also recognize the main open issues in this research area. Based on the results of this SLR, we conclude that despite the potential and continuous evolution for using these solutions as medical decision support tools, this research area needs more work to overcome technology and methodological rigor issues to adopt proposed solutions in real clinical settings.

The Use of Passive Smartphone Data to Monitor Anxiety and Depression Among College Students in Real-World Settings: Protocol for a Systematic Review

BACKGROUND

College students are particularly at risk of depression and anxiety. These disorders have a serious impact on public health and affect patients' daily lives. The potential for using smartphones to monitor these mental conditions, providing passively collected physiological and behavioral data, has been reported among the general population. However, research on the use of passive smartphone data to monitor anxiety and depression among specific populations of college students has never been reviewed.

OBJECTIVE

This review's objectives are (1) to provide an overview of the use of passive smartphone data to monitor depression and anxiety among college students, given their specific type of smartphone use and living setting, and (2) to evaluate the different methods used to assess those smartphone data, including their strengths and limitations.

METHODS

This review will follow the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Two independent investigators will review English-language, full-text, peer-reviewed papers extracted from PubMed and Web of Science that measure passive smartphone data and levels of depression or anxiety among college students. A preliminary search was conducted in February 2022 as a proof of concept.

RESULTS

Our preliminary search identified 115 original articles, 8 of which met our eligibility criteria. Our planned full study will include an article selection flowchart, tables, and figures representing the main information extracted on the use of passive smartphone data to monitor anxiety and depression among college students.

CONCLUSIONS

The planned review will summarize the published research on using passive smartphone data to monitor anxiety and depression among college students. The review aims to better understand whether and how passive smartphone data are associated with indicators of depression and anxiety among college students. This could be valuable in order to provide a digital solution for monitoring mental health issues in this specific population by enabling easier identification and follow-up of the patients.

TRIAL REGISTRATION

PROSPERO CRD42022316263; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=316263.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/38785.

Personalised depression forecasting using mobile sensor data and ecological momentary assessment

INTRODUCTION

Digital health interventions are an effective way to treat depression, but it is still largely unclear how patients' individual symptoms evolve dynamically during such treatments. Data-driven forecasts of depressive symptoms would allow to greatly improve the personalisation of treatments. In current forecasting approaches, models are often trained on an entire population, resulting in a general model that works overall, but does not translate well to each individual in clinically heterogeneous, real-world populations. Model fairness across patient subgroups is also frequently overlooked. Personalised models tailored to the individual patient may therefore be promising.

METHODS

We investigate different personalisation strategies using transfer learning, subgroup models, as well as subject-dependent standardisation on a newly-collected, longitudinal dataset of depression patients undergoing treatment with a digital intervention ( N = 65 patients recruited). Both passive mobile sensor data as well as ecological momentary assessments were available for modelling. We evaluated the models' ability to predict symptoms of depression (Patient Health Questionnaire-2; PHQ-2) at the end of each day, and to forecast symptoms of the next day.

RESULTS

In our experiments, we achieve a best mean-absolute-error (MAE) of 0.801 (25% improvement) for predicting PHQ-2 values at the end of the day with subject-dependent standardisation compared to a non-personalised baseline ( MAE = 1.062 ). For one day ahead-forecasting, we can improve the baseline of 1.539 by 12 % to a MAE of 1.349 using a transfer learning approach with shared common layers. In addition, personalisation leads to fairer models at group-level.

DISCUSSION

Our results suggest that personalisation using subject-dependent standardisation and transfer learning can improve predictions and forecasts, respectively, of depressive symptoms in participants of a digital depression intervention. We discuss technical and clinical limitations of this approach, avenues for future investigations, and how personalised machine learning architectures may be implemented to improve existing digital interventions for depression.

A common factors perspective on mindfulness-based interventions

Mindfulness-based interventions (MBIs) have entered mainstream Western culture in the past four decades. There are now dozens of MBIs with varying degrees of empirical support and a variety of mindfulness-specific psychological mechanisms have been proposed to account for the beneficial effects of MBIs. Although it has long been acknowledged that non-specific or common factors might contribute to MBI efficacy, relatively little empirical work has directly investigated these aspects. In this Perspective, I suggest that situating MBIs within the broader psychotherapy research literature and emphasizing the commonalities rather than differences between MBIs and other treatments might help guide future MBI research. To that end, I summarize the evidence for MBI efficacy and several MBI-specific psychological mechanisms, contextualize MBI findings within the broader psychotherapy literature from a common factors perspective, and propose suggestions for future research based on innovations and challenges occurring within psychotherapy research.

Personal mental health navigator: Harnessing the power of data, personal models, and health cybernetics to promote psychological well-being

Current digital mental healthcare solutions conventionally take on a reactive approach, requiring individuals to self-monitor and document existing symptoms. These solutions are unable to provide comprehensive, wrap-around, customized treatments that capture an individual’s holistic mental health model as it unfolds over time. Recognizing that each individual requires personally tailored mental health treatment, we introduce the notion of Personalized Mental Health Navigation (MHN): a cybernetic goal-based system that deploys a continuous loop of monitoring, estimation, and guidance to steer the individual towards mental flourishing. We present the core components of MHN that are premised on the importance of addressing an individual’s personal mental health state. Moreover, we provide an overview of the existing physical health navigation systems and highlight the requirements and challenges of deploying the navigational approach to the mental health domain.

Exploring the digital footprint of depression: a PRISMA systematic literature review of the empirical evidence

BACKGROUND

This PRISMA systematic literature review examined the use of digital data collection methods (including ecological momentary assessment [EMA], experience sampling method [ESM], digital biomarkers, passive sensing, mobile sensing, ambulatory assessment, and time-series analysis), emphasizing on digital phenotyping (DP) to study depression. DP is defined as the use of digital data to profile health information objectively.

AIMS

Four distinct yet interrelated goals underpin this study: (a) to identify empirical research examining the use of DP to study depression; (b) to describe the different methods and technology employed; (c) to integrate the evidence regarding the efficacy of digital data in the examination, diagnosis, and monitoring of depression and (d) to clarify DP definitions and digital mental health records terminology.

RESULTS

Overall, 118 studies were assessed as eligible. Considering the terms employed, "EMA", "ESM", and "DP" were the most predominant. A variety of DP data sources were reported, including voice, language, keyboard typing kinematics, mobile phone calls and texts, geocoded activity, actigraphy sensor-related recordings (i.e., steps, sleep, circadian rhythm), and self-reported apps' information. Reviewed studies employed subjectively and objectively recorded digital data in combination with interviews and psychometric scales.

CONCLUSIONS

Findings suggest links between a person's digital records and depression. Future research recommendations include (a) deriving consensus regarding the DP definition and (b) expanding the literature to consider a person's broader contextual and developmental circumstances in relation to their digital data/records.

Some Recommendations on the Use of Daily Life Methods in Affective Science

Real-world emotions are often more vivid, personally meaningful, and consequential than those evoked in the lab. Therefore, studying emotions in daily life is essential to test theories, discover new phenomena, and understand healthy emotional functioning; in short, to move affective science forward. The past decades have seen a surge of research using daily diary, experience sampling, or ecological momentary assessment methods to study emotional phenomena in daily life. In this paper, we will share some of the insights we have gained from our collective experience applying such daily life methods to study everyday affective processes. We highlight what we see as important considerations and caveats involved in using these methods and formulate recommendations to improve their use in future research. These insights focus on the importance of (i) theory and hypothesis-testing; (ii) measurement; (iii) timescale; and (iv) context, when studying emotions in their natural habitat.

Association between Daily Pattern of Physical Activity and Depression: A Systematic Review

Recent research suggested that daily pattern of physical activity (PA) may have an important association with depression, but findings are limited and contradictory. Our aim was to conduct a systematic review of the literature to summarize the literature evidence on the association between timing of PA and depression. A comprehensive search of PubMed/Medline and Scopus databases has been performed, and a total of five manuscripts have been thoroughly reviewed. The performed descriptive analysis shows lower levels of PA among individuals with depression or depressive symptoms, although evidence on the 24 h pattern of PA and depression is limited. An interesting finding is the association between lower PA during the morning, higher PA late in the evening (night), and depression or depressive symptoms. However, definitive conclusions could not be drawn due to the observational nature of the studies, their limited number, the high heterogeneity in the sample populations, and the studies’ differing outcome definitions and exposure assessments. Future studies considering not only the level of PA but also its daily variability might be important to further explore this novel area of research.

Digital phenotyping in psychiatry: A scoping review

BACKGROUND

Digital phenotyping has been defined as the moment-by-moment assessment of an illness state through digital means, promising objective, quantifiable data on psychiatric patients' conditions, and could potentially improve diagnosis and management of mental illness. As it is a rapidly growing field, it is to be expected that new literature is being published frequently.

OBJECTIVE

We conducted this scoping review to assess the current state of literature on digital phenotyping and offer some discussion on the current trends and future direction of this area of research.

METHODS

We searched four databases, PubMed, Ovid MEDLINE, PsycINFO and Web of Science, from inception to August 25th, 2021. We included studies written in English that 1) investigated or applied their findings to diagnose psychiatric disorders and 2) utilized passive sensing for management or diagnosis. Protocols were excluded. A narrative synthesis approach was used, due to the heterogeneity and variability in outcomes and outcome types reported.

RESULTS

Of 10506 unique records identified, we included a total of 107 articles. The number of published studies has increased over tenfold from 2 in 2014 to 28 in 2020, illustrating the field's rapid growth. However, a significant proportion of these (49% of all studies and 87% of primary studies) were proof of concept, pilot or correlational studies examining digital phenotyping's potential. Most (62%) of the primary studies published evaluated individuals with depression (21%), BD (18%) and SZ (23%) (Appendix 1).

CONCLUSION

There is promise shown in certain domains of data and their clinical relevance, which have yet to be fully elucidated. A consensus has yet to be reached on the best methods of data collection and processing, and more multidisciplinary collaboration between physicians and other fields is needed to unlock the full potential of digital phenotyping and allow for statistically powerful clinical trials to prove clinical utility.

Using Smartphone Sensor Paradata and Personalized Machine Learning Models to Infer Participants' Well-being: Ecological Momentary Assessment

Background

Sensors embedded in smartphones allow for the passive momentary quantification of people’s states in the context of their daily lives in real time. Such data could be useful for alleviating the burden of ecological momentary assessments and increasing utility in clinical assessments. Despite existing research on using passive sensor data to assess participants’ moment-to-moment states and activity levels, only limited research has investigated temporally linking sensor assessment and self-reported assessment to further integrate the 2 methodologies.

Objective

We investigated whether sparse movement-related sensor data can be used to train machine learning models that are able to infer states of individuals’ work-related rumination, fatigue, mood, arousal, life engagement, and sleep quality. Sensor data were only collected while the participants filled out the questionnaires on their smartphones.

Methods

We trained personalized machine learning models on data from employees (N=158) who participated in a 3-week ecological momentary assessment study.

Results

The results suggested that passive smartphone sensor data paired with personalized machine learning models can be used to infer individuals’ self-reported states at later measurement occasions. The mean R² was approximately 0.31 (SD 0.29), and more than half of the participants (119/158, 75.3%) had an R² of ≥0.18. Accuracy was only slightly attenuated compared with earlier studies and ranged from 38.41% to 51.38%.

Conclusions

Personalized machine learning models and temporally linked passive sensing data have the capability to infer a sizable proportion of variance in individuals’ daily self-reported states. Further research is needed to investigate factors that affect the accuracy and reliability of the inference.

Detecting Mental Health Behaviors Using Mobile Interactions: Exploratory Study Focusing on Binge Eating

BACKGROUND

Binge eating is a subjective loss of control while eating, which leads to the consumption of large amounts of food. It can cause significant emotional distress and is often accompanied by purging behaviors (eg, meal skipping, overexercising, or vomiting).

OBJECTIVE

The aim of this study was to explore the potential of mobile sensing to detect indicators of binge-eating episodes, with a view toward informing the design of future context-aware mobile interventions.

METHODS

This study was conducted in 2 stages. The first involved the development of the DeMMI (Detecting Mental health behaviors using Mobile Interactions) app. As part of this, we conducted a consultation session to explore whether the types of sensor data we were proposing to capture were useful and appropriate, as well as to gather feedback on some specific app features relating to self-reporting. The second stage involved conducting a 6-week period of data collection with 10 participants experiencing binge eating (logging both their mood and episodes of binge eating) and 10 comparison participants (logging only mood). An optional interview was conducted after the study, which discussed their experience using the app, and 8 participants (n=3, 38% binge eating and n=5, 63% comparisons) consented.

RESULTS

The findings showed unique differences in the types of sensor data that were triangulated with the individuals' episodes (with nearby Bluetooth devices, screen and app use features, mobility features, and mood scores showing relevance). Participants had a largely positive opinion about the app, its unobtrusive role, and its ease of use. Interacting with the app increased participants' awareness of and reflection on their mood and phone usage patterns. Moreover, they expressed no privacy concerns as these were alleviated by the study information sheet.

CONCLUSIONS

This study contributes a series of recommendations for future studies wishing to scale our approach and for the design of bespoke mobile interventions to support this population.

Concurrent and prospective associations between fitbit wearable-derived RDoC arousal and regulatory constructs and adolescent internalizing symptoms

BACKGROUND

Adolescence is characterized by alterations in biobehavioral functioning, during which individuals are at heightened risk for onset of psychopathology, particularly internalizing disorders. Researchers have proposed using digital technologies to index daily biobehavioral functioning, yet there is a dearth of research examining how wearable metrics are associated with mental health.

METHODS

We preregistered analyses using the Adolescent Brain Cognitive Development Study dataset using wearable data collection in 5,686 adolescents (123,862 person-days or 2,972,688 person-hours) to determine whether wearable indices of resting heart rate (RHR), step count, and sleep duration and variability in these measures were cross-sectionally associated with internalizing symptomatology. All models were also run controlling for age, sex, body mass index, socioeconomic status, and race. We then performed prospective analyses on a subset of this sample (n = 143) across 25 months that had Fitbit data available at baseline and follow-up in order to explore directionality of effects.

RESULTS

Cross-sectional analyses revealed a small, yet significant, effect size (R²  = .053) that higher RHR, lower step count and step count variability, and greater variability in sleep duration were associated with greater internalizing symptoms. Cross-lagged panel model analysis revealed that there were no prospective associations between wearable variables and internalizing symptoms (partial R²  = .026), but greater internalizing symptoms and higher RHR predicted lower step count 25 months later (partial R²  = .010), while higher RHR also predicted lower step count variability 25 months later (partial R²  = .008).

CONCLUSIONS

Findings indicate that wearable indices concurrently associate with internalizing symptoms during early adolescence, while a larger sample size is likely required to accurately assess prospective or directional effects between wearable indices and mental health. Future research should capitalize on the temporal resolution provided by wearable devices to determine the intensive longitudinal relations between biobehavioral risk factors and acute changes in mental health.

Sensing Apps and Public Data Sets for Digital Phenotyping of Mental Health: Systematic Review

Background

Mental disorders are normally diagnosed exclusively on the basis of symptoms, which are identified from patients’ interviews and self-reported experiences. To make mental health diagnoses and monitoring more objective, different solutions have been proposed such as digital phenotyping of mental health (DPMH), which can expand the ability to identify and monitor health conditions based on the interactions of people with digital technologies.

Objective

This article aims to identify and characterize the sensing applications and public data sets for DPMH from a technical perspective.

Methods

We performed a systematic review of scientific literature and data sets. We searched 8 digital libraries and 20 data set repositories to find results that met the selection criteria. We conducted a data extraction process from the selected articles and data sets. For this purpose, a form was designed to extract relevant information, thus enabling us to answer the research questions and identify open issues and research trends.

Results

A total of 31 sensing apps and 8 data sets were identified and reviewed. Sensing apps explore different context data sources (eg, positioning, inertial, ambient) to support DPMH studies. These apps are designed to analyze and process collected data to classify (n=11) and predict (n=6) mental states/disorders, and also to investigate existing correlations between context data and mental states/disorders (n=6). Moreover, general-purpose sensing apps are developed to focus only on contextual data collection (n=9). The reviewed data sets contain context data that model different aspects of human behavior, such as sociability, mood, physical activity, sleep, with some also being multimodal.

Conclusions

This systematic review provides in-depth analysis regarding solutions for DPMH. Results show growth in proposals for DPMH sensing apps in recent years, as opposed to a scarcity of public data sets. The review shows that there are features that can be measured on smart devices that can act as proxies for mental status and well-being; however, it should be noted that the combined evidence for high-quality features for mental states remains limited. DPMH presents a great perspective for future research, mainly to reach the needed maturity for applications in clinical settings.

Digital biomarkers of anxiety disorder symptom changes: Personalized deep learning models using smartphone sensors accurately predict anxiety symptoms from ecological momentary assessments

Smartphones are capable of passively capturing persons' social interactions, movement patterns, physiological activation, and physical environment. Nevertheless, little research has examined whether momentary anxiety symptoms can be accurately assessed using these methodologies. In this research, we utilize smartphone sensors and personalized deep learning models to predict future anxiety symptoms among a sample reporting clinical anxiety disorder symptoms. Participants (N = 32) with generalized anxiety disorder and/or social anxiety disorder (based on self-report) installed a smartphone application and completed ecological momentary assessment symptoms assessing their anxiety and avoidance symptoms hourly for the course of one week (T = 2007 assessments). During the same period, the smartphone app collected information about physiological activation (heart rate and heart rate variability), exposure to light, social contact, and GPS location. GPS locations were coded to reveal the type of location and the weather information. Personalized deep learning models using the smartphone sensor data were capable of predicting the majority of total variation in anxiety symptoms (R² = 0.748) and predicting a large proportion of within-person variation at the hour-by-hour level (mean R² = 0.385). These results suggest that personalized deep learning models using smartphone sensor data are capable of accurately predicting future anxiety disorder symptom changes.

Using Smartphone App Use and Lagged-Ensemble Machine Learning for the Prediction of Work Fatigue and Boredom

INTRO

As smartphone usage becomes increasingly prevalent in the workplace, the physical and psychological implications of this behavior warrant consideration. Recent research has investigated associations between workplace smartphone use and fatigue and boredom, yet findings are not conclusive.

METHODS

To build off recent efforts, we applied an ensemble machine learning model on a previously published dataset of N = 83 graduate students in the Netherlands to predict work boredom and fatigue from passively collected smartphone app use information. Using time-based feature engineering and lagged variations of the data to train, validate, and test idiographic models, we evaluated the efficacy of a lagged-ensemble predictive paradigm on sparse temporal data. Moreover, we probed the relative importance of both derived app use variables and lags within this predictive framework.

RESULTS

The ability to predict fatigue and boredom trajectories from app use information was heterogeneous and highly person-specific. Idiographic modeling reflected moderate to high correlative capacity (r > 0.4) in 47% of participants for fatigue and 24% for boredom, with better overall performance in the fatigue prediction task. App use relating to duration, communication, and patterns of use frequency were among the most important features driving predictions across lags, with longer lags contributing more heavily to final ensemble predictions compared with shorter ones.

CONCLUSION

A lag- specific ensemble predictive paradigm is a promising approach to leveraging high-dimensional app use behavioral data for the prediction of work fatigue and boredom. Future research will benefit from evaluating associations on densely collected data across longer time scales.

Digital phenotype of mood disorders: A conceptual and critical review

BACKGROUND

Mood disorders are commonly diagnosed and staged using clinical features that rely merely on subjective data. The concept of digital phenotyping is based on the idea that collecting real-time markers of human behavior allows us to determine the digital signature of a pathology. This strategy assumes that behaviors are quantifiable from data extracted and analyzed through digital sensors, wearable devices, or smartphones. That concept could bring a shift in the diagnosis of mood disorders, introducing for the first time additional examinations on psychiatric routine care.

OBJECTIVE

The main objective of this review was to propose a conceptual and critical review of the literature regarding the theoretical and technical principles of the digital phenotypes applied to mood disorders.

METHODS

We conducted a review of the literature by updating a previous article and querying the PubMed database between February 2017 and November 2021 on titles with relevant keywords regarding digital phenotyping, mood disorders and artificial intelligence.

RESULTS

Out of 884 articles included for evaluation, 45 articles were taken into account and classified by data source (multimodal, actigraphy, ECG, smartphone use, voice analysis, or body temperature). For depressive episodes, the main finding is a decrease in terms of functional and biological parameters [decrease in activities and walking, decrease in the number of calls and SMS messages, decrease in temperature and heart rate variability (HRV)], while the manic phase produces the reverse phenomenon (increase in activities, number of calls and HRV).

CONCLUSION

The various studies presented support the potential interest in digital phenotyping to computerize the clinical characteristics of mood disorders.

Wearable, Environmental, and Smartphone-Based Passive Sensing for Mental Health Monitoring

Collecting and analyzing data from sensors embedded in the context of daily life has been widely employed for the monitoring of mental health. Variations in parameters such as movement, sleep duration, heart rate, electrocardiogram, skin temperature, etc., are often associated with psychiatric disorders. Namely, accelerometer data, microphone, and call logs can be utilized to identify voice features and social activities indicative of depressive symptoms, and physiological factors such as heart rate and skin conductance can be used to detect stress and anxiety disorders. Therefore, a wide range of devices comprising a variety of sensors have been developed to capture these physiological and behavioral data and translate them into phenotypes and states related to mental health. Such systems aim to identify behaviors that are the consequence of an underlying physiological alteration, and hence, the raw sensor data are captured and converted into features that are used to define behavioral markers, often through machine learning. However, due to the complexity of passive data, these relationships are not simple and need to be well-established. Furthermore, parameters such as intrapersonal and interpersonal differences need to be considered when interpreting the data. Altogether, combining practical mobile and wearable systems with the right data analysis algorithms can provide a useful tool for the monitoring and management of mental disorders. The current review aims to comprehensively present and critically discuss all available smartphone-based, wearable, and environmental sensors for detecting such parameters in relation to the treatment and/or management of the most common mental health conditions.

Just-in-Time Adaptive Mechanisms of Popular Mobile Apps for Individuals With Depression: Systematic App Search and Literature Review

BACKGROUND

The number of smartphone apps that focus on the prevention, diagnosis, and treatment of depression is increasing. A promising approach to increase the effectiveness of the apps while reducing the individual's burden is the use of just-in-time adaptive intervention (JITAI) mechanisms. JITAIs are designed to improve the effectiveness of the intervention and reduce the burden on the person using the intervention by providing the right type of support at the right time. The right type of support and the right time are determined by measuring the state of vulnerability and the state of receptivity, respectively.

OBJECTIVE

The aim of this study is to systematically assess the use of JITAI mechanisms in popular apps for individuals with depression.

METHODS

We systematically searched for apps addressing depression in the Apple App Store and Google Play Store, as well as in curated lists from the Anxiety and Depression Association of America, the United Kingdom National Health Service, and the American Psychological Association in August 2020. The relevant apps were ranked according to the number of reviews (Apple App Store) or downloads (Google Play Store). For each app, 2 authors separately reviewed all publications concerning the app found within scientific databases (PubMed, Cochrane Register of Controlled Trials, PsycINFO, Google Scholar, IEEE Xplore, Web of Science, ACM Portal, and Science Direct), publications cited on the app's website, information on the app's website, and the app itself. All types of measurements (eg, open questions, closed questions, and device analytics) found in the apps were recorded and reviewed.

RESULTS

None of the 28 reviewed apps used JITAI mechanisms to tailor content to situations, states, or individuals. Of the 28 apps, 3 (11%) did not use any measurements, 20 (71%) exclusively used self-reports that were insufficient to leverage the full potential of the JITAIs, and the 5 (18%) apps using self-reports and passive measurements used them as progress or task indicators only. Although 34% (23/68) of the reviewed publications investigated the effectiveness of the apps and 21% (14/68) investigated their efficacy, no publication mentioned or evaluated JITAI mechanisms.

CONCLUSIONS

Promising JITAI mechanisms have not yet been translated into mainstream depression apps. Although the wide range of passive measurements available from smartphones were rarely used, self-reported outcomes were used by 71% (20/28) of the apps. However, in both cases, the measured outcomes were not used to tailor content and timing along a state of vulnerability or receptivity. Owing to this lack of tailoring to individual, state, or situation, we argue that the apps cannot be considered JITAIs. The lack of publications investigating whether JITAI mechanisms lead to an increase in the effectiveness or efficacy of the apps highlights the need for further research, especially in real-world apps.

Digital Phenotyping in Livestock Farming

Currently, large volumes of data are being collected on farms using multimodal sensor technologies. These sensors measure the activity, housing conditions, feed intake, and health of farm animals. With traditional methods, the data from farm animals and their environment can be collected intermittently. However, with the advancement of wearable and non-invasive sensing tools, these measurements can be made in real-time for continuous quantitation relating to clinical biomarkers, resilience indicators, and behavioral predictors. The digital phenotyping of humans has drawn enormous attention recently due to its medical significance, but much research is still needed for the digital phenotyping of farm animals. Implications from human studies show great promise for the application of digital phenotyping technology in modern livestock farming, but these technologies must be directly applied to animals to understand their true capacities. Due to species-specific traits, certain technologies required to assess phenotypes need to be tailored efficiently and accurately. Such devices allow for the collection of information that can better inform farmers on aspects of animal welfare and production that need improvement. By explicitly addressing farm animals' individual physiological and mental (affective states) needs, sensor-based digital phenotyping has the potential to serve as an effective intervention platform. Future research is warranted for the design and development of digital phenotyping technology platforms that create shared data standards, metrics, and repositories.

Personalized machine learning of depressed mood using wearables

Depression is a multifaceted illness with large interindividual variability in clinical response to treatment. In the era of digital medicine and precision therapeutics, new personalized treatment approaches are warranted for depression. Here, we use a combination of longitudinal ecological momentary assessments of depression, neurocognitive sampling synchronized with electroencephalography, and lifestyle data from wearables to generate individualized predictions of depressed mood over a 1-month time period. This study, thus, develops a systematic pipeline for N-of-1 personalized modeling of depression using multiple modalities of data. In the models, we integrate seven types of supervised machine learning (ML) approaches for each individual, including ensemble learning and regression-based methods. All models were verified using fourfold nested cross-validation. The best-fit as benchmarked by the lowest mean absolute percentage error, was obtained by a different type of ML model for each individual, demonstrating that there is no one-size-fits-all strategy. The voting regressor, which is a composite strategy across ML models, was best performing on-average across subjects. However, the individually selected best-fit models still showed significantly less error than the voting regressor performance across subjects. For each individual's best-fit personalized model, we further extracted top-feature predictors using Shapley statistics. Shapley values revealed distinct feature determinants of depression over time for each person ranging from co-morbid anxiety, to physical exercise, diet, momentary stress and breathing performance, sleep times, and neurocognition. In future, these personalized features can serve as targets for a personalized ML-guided, multimodal treatment strategy for depression.

Using artificial intelligence and longitudinal location data to differentiate persons who develop posttraumatic stress disorder following childhood trauma

Post-traumatic stress disorder (PTSD) is characterized by complex, heterogeneous symptomology, thus detection outside traditional clinical contexts is difficult. Fortunately, advances in mobile technology, passive sensing, and analytics offer promising avenues for research and development. The present study examined the ability to utilize Global Positioning System (GPS) data, derived passively from a smartphone across seven days, to detect PTSD diagnostic status among a cohort (N = 185) of high-risk, previously traumatized women. Using daily time spent away and maximum distance traveled from home as a basis for model feature engineering, the results suggested that diagnostic group status can be predicted out-of-fold with high performance (AUC = 0.816, balanced sensitivity = 0.743, balanced specificity = 0.8, balanced accuracy = 0.771). Results further implicate the potential utility of GPS information as a digital biomarker of the PTSD behavioral repertoire. Future PTSD research will benefit from application of GPS data within larger, more diverse populations.

Performance Assessment of Certain Machine Learning Models for Predicting the Major Depressive Disorder among IT Professionals during Pandemic times

Major depressive disorder (MDD) is the most common mental disorder in the present day as all individuals' lives, irrespective of being employed or unemployed, is going through the depression phase at least once in their lifetime. In simple terms, it is a mood disturbance that can persist for an individual for more than a few weeks to months. In MDD, in most cases, the individuals do not consult a professional, and even if being consulted, the results are not significant as the individuals find it challenging to identify whether they are depressed or not. Depression, most of the time, cooccurs with anxiety and leads to suicide in few cases, among the employees, who are about to handle the pressure at work and home and mostly unnoticing such problems. This is why this work aims to analyze the IT employees who are mostly working with targets. The artificial neural network, which is modeled loosely like the brain, has proved in recent days that it can perform better than most of the classification algorithms. This study has implemented the multilayered neural perceptron and experimented with the backpropagation technique over the data samples collected from IT professionals. This study aims to develop a model that can classify depressed individuals from those who are not depressed effectively with the data collected from them manually and through sensors. The results show that deep-MLP with backpropagation outperforms other machine learning-based models for effective classification.

Ecological momentary interventions for mental health: A scoping review

BACKGROUND

The development of mobile computing technology has enabled the delivery of psychological interventions while people go about their everyday lives. The original visions of the potential of these "ecological momentary interventions" were presented over a decade ago, and the widespread adoption of smartphones in the intervening years has led to a variety of research studies exploring the feasibility of these aspirations. However, there is a dearth of research describing the different dimensions, characteristics, and features of these interventions, as constructed.

OBJECTIVE

To provide an overview of the definitions given for "ecological momentary interventions" in the treatment of common mental health disorders, and describe the set of technological and interaction possibilities which have been used in the design of these interventions.

METHODS

A systematic search identified relevant literature published between 2009 and 2020 in the PubMed, PsycInfo, and ACM Guide to the Computing Literature databases. Following screening, data were extracted from eligible articles using a standardized extraction worksheet. Selected articles were then thematically categorized.

RESULTS

The search identified 583 articles of which 64 met the inclusion criteria. The interventions target a range of mental health problems, with diverse aims, intervention designs and evaluation approaches. The studies employed a variety of features for intervention delivery, but recent research is overwhelmingly comprised of studies based on smartphone apps (30 of 42 papers that described an intervention). Twenty two studies employed sensors for the collection of data in order to provide just-in-time support or predict psychological states.

CONCLUSIONS

With the shift towards smartphone apps, the vision for EMIs has begun to be realised. Recent years have seen increased exploration of the use of sensors and machine learning, but the role of humans in the delivery of EMI is also varied. The variety of capabilities exhibited by EMIs motivates development of a more precise vocabulary for capturing both automatic and human tailoring of these interventions.

Assessing the Mental Health of Emerging Adults Through a Mental Health App: Protocol for a Prospective Pilot Study

BACKGROUND

Individuals can experience different manifestations of the same psychological disorder. This underscores the need for a personalized model approach in the study of psychopathology. Emerging adulthood is a developmental phase wherein individuals are especially vulnerable to psychopathology. Given their exposure to repeated stressors and disruptions in routine, the emerging adult population is worthy of investigation.

OBJECTIVE

In our prospective study, we aim to conduct multimodal assessments to determine the feasibility of an individualized approach for understanding the contextual factors of changes in daily affect, sleep, physiology, and activity. In other words, we aim to use event mining to predict changes in mental health.

METHODS

We expect to have a final sample size of 20 participants. Recruited participants will be monitored for a period of time (ie, between 3 and 12 months). Participants will download the Personicle app on their smartphone to track their activities (eg, home events and cycling). They will also be given wearable sensor devices (ie, devices that monitor sleep, physiology, and physical activity), which are to be worn continuously. Participants will be asked to report on their daily moods and provide open-ended text responses on a weekly basis. Participants will be given a battery of questionnaires every 3 months.

RESULTS

Our study has been approved by an institutional review board. The study is currently in the data collection phase. Due to the COVID-19 pandemic, the study was adjusted to allow for remote data collection and COVID-19-related stress assessments.

CONCLUSIONS

Our study will help advance research on individualized approaches to understanding health and well-being through multimodal systems. Our study will also demonstrate the benefit of using individualized approaches to study interrelations among stress, social relationships, technology, and mental health.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/25775.

Deep learning paired with wearable passive sensing data predicts deterioration in anxiety disorder symptoms across 17-18 years

BACKGROUND

Recent studies have demonstrated that passive smartphone and wearable sensor data collected throughout daily life can predict anxiety symptoms cross-sectionally. However, to date, no research has demonstrated the capacity for these digital biomarkers to predict long-term prognosis.

METHODS

We utilized deep learning models based on wearable sensor technology to predict long-term (17-18-year) deterioration in generalized anxiety disorder and panic disorder symptoms from actigraphy data on daytime movement and nighttime sleeping patterns. As part of Midlife in the United States (MIDUS), a national longitudinal study of health and well-being, subjects (N = 265) (i) completed a phone-based interview that assessed generalized anxiety disorder and panic disorder symptoms at enrollment, (ii) participated in a one-week actigraphy study 9-14 years later, and (iii) completed a long-term follow-up, phone-based interview to quantify generalized anxiety disorder and panic disorder symptoms 17-18 years from initial enrollment. A deep auto-encoder paired with a multi-layered ensemble deep learning model was leveraged to predict whether participants experienced increased anxiety disorder symptoms across this 17-18 year period.

RESULTS

Out-of-sample cross-validated results suggested that wearable movement data could significantly predict which individuals would experience symptom deterioration (AUC = 0.696, CI [0.598, 0.793], 84.6% sensitivity, 52.7% specificity, balanced accuracy = 68.7%).

CONCLUSIONS

Passive wearable actigraphy data could be utilized to predict long-term deterioration of anxiety disorder symptoms. Future studies should examine whether these methods could be implemented to prevent deterioration of anxiety disorder symptoms.

The dynamics, processes, mechanisms, and functioning of personality: An overview of the field

Personality psychology has long focused on structural trait models, but it can also offer a rich understanding of the dynamics, processes, mechanisms, and functioning of individual differences or entire persons. The field of personality dynamics, which works towards such an understanding, has experienced a renaissance in the last two decades. This review article seeks to act as a primer of that field. It covers its historical roots, summarizes current research strands - along with their theoretical backbones and methodologies - in an accessible way, and sketches some considerations for the future. In doing so, we introduce relevant concepts, give an overview of different topics and phenomena subsumed under the broad umbrella term 'dynamics', and highlight the interdisciplinarity as well as applied relevance of the field. We hope this article can serve as a useful overview for scholars within and outside of personality psychology who are interested in the dynamic nature of human behaviour and experience.

Disentangling Trait-Like Between-Individual vs. State-Like Within-Individual Effects in Studying the Mechanisms of Change in CBT

Hofmann et al. argued that "[w]hile the clinical field has produced a dizzying number of treatment models and treatment protocols for virtually every psychiatric and psychological problem imaginable, increases in understanding of the processes of change in psychotherapy has been slow to arrive." We propose that one of the reasons for the slow progress is that prior psychotherapy research conflates trait-like and state-like components of mechanisms of change. Trait-like components can serve as prescriptive or prognostic variables, whereas state-like components reflect within-client processes of change, and may highlight active ingredients of successful treatment. Distinguishing between the two is essential for clarifying the underlying processes of change in psychotherapy, and ultimately identifying empirically-derived individualized treatment targets. We review studies that implement methodological and statistical approaches for disentangling the two. These studies clarified particular mechanisms of change that may operate in a given treatment, highlighted differences in the processes of change between different treatments, and explored the within-individual interplay between different mechanisms of change during treatment. Examples include studies investigating the therapeutic role of behavioral, cognitive, and interpersonal skills, as well as emotional processing. We conclude with suggestions for future research, including attention to diversity, improved measurement to facilitate a reliable and valid estimation of trait-like and state-like components, the use of appropriate statistical approaches to adequately disentangle the two components, integration of theory-driven and data-driven methods of analysis, and the need to experimentally manipulate the state-like changes in a given mechanism of change to strengthen causal inferences.

The Use of Digitally Assessed Stress Levels to Model Change Processes in CBT - A Feasibility Study on Seven Case Examples

In psychotherapy research, the measurement of treatment processes and outcome are predominantly based on self-reports. However, given new technological developments, other potential sources can be considered to improve measurements. In a feasibility study, we examined whether Ecological Momentary Assessments (EMA) using digital phenotyping (stress level) can be a valuable tool to investigate change processes during cognitive behavioral therapy (CBT). Seven outpatients undergoing psychological treatment were assessed using EMA. Continuous stress levels (heart rate variability) were assessed via fitness trackers (Garmin) every 3 min over a 2-week time period (6,720 measurements per patient). Time-varying change point autoregressive (TVCP-AR) models were employed to detect both gradual and abrupt changes in stress levels. Results for seven case examples indicate differential patterns of change processes in stress. More precisely, inertia of stress level changed gradually over time in one of the participants, whereas the other participants showed both gradual and abrupt changes. This feasibility study demonstrates that intensive longitudinal assessments enriched by digitally assessed stress levels have the potential to investigate intra- and interindividual differences in treatment change processes and their relations to treatment outcome. Further, implementation issues and implications for future research and developments using digital phenotyping are discussed.

Use of Passive Sensing in Psychotherapy Studies in Late Life: A Pilot Example, Opportunities and Challenges

Late-life depression is heterogenous and patients vary in disease course over time. Most psychotherapy studies measure activity levels and symptoms solely using self-report scales, administered periodically. These scales may not capture granular changes during treatment. We introduce the potential utility of passive sensing data collected with smartphone to assess fluctuations in daily functioning in real time during psychotherapy for late life depression in elder abuse victims. To our knowledge, this is the first investigation of passive sensing among depressed elder abuse victims. We present data from three victims who received a 9-week intervention as part of a pilot randomized controlled trial and showed a significant decrease in depressive symptoms (50% reduction). Using a smartphone, we tracked participants' daily number of smartphone unlocks, time spent at home, time spent in conversation, and step count over treatment. Independent assessment of depressive symptoms and behavioral activation were collected at intake, Weeks 6 and 9. Data revealed patient-level fluctuations in activity level over treatment, corresponding with self-reported behavioral activation. We demonstrate how passive sensing data could expand our understanding of heterogenous presentations of late-life depression among elder abuse. We illustrate how trajectories of change in activity levels as measured with passive sensing and subjective measures can be tracked concurrently over time. We outline challenges and potential solutions for application of passive sensing data collection in future studies with larger samples using novel advanced statistical modeling, such as artificial intelligence algorithms.