Digital Phenotyping for Mental Health: Reviewing the Challenges of Using Data to Monitor and Predict Mental Health Problems

Digital phenotyping for mental health based on data analytics: A systematic literature review

Even though mental health is a human right, mental disorders still affect millions of people worldwide. Untreated and undertreated mental health conditions may lead to suicide, which generates more than 700,000 deaths annually around the world. The broad adoption of smartphones and wearable devices allowed the recording and analysis of human behaviors in digital devices, which might reveal mental health symptoms. This analysis constitutes digital phenotyping research, referring to frequent and constant measurement of human phenotypes in situ based on data from smartphones and other personal digital devices. Therefore, this article presents a systematic literature review providing a computer science view on data analytics for digital phenotyping in mental health. This study reviewed 5,422 articles from ten academic databases published up to September 2024, generating a final list of 74 studies. The investigated databases are ACM, IEEE Xplore, PsycArticles, PsycInfo, Pubmed, Science Direct, Scopus, Springer, Web of Science, and Wiley. We investigated ten research questions, considering explored data, employed devices, and techniques for data analysis. This review also organizes the application domains and mental health conditions, data analytics techniques, and current research challenges. This study found a growing research interest in digital phenotyping for mental health in recent years. Current approaches still present a high dependence on self-reported measures of mental health status, but there is evidence of the employment of smartphones for leveraging passive data collection. Traditional machine learning techniques are the main explored strategies for analyzing the large amount of collected data. In this regard, published approaches deeply focused on data analysis, generating opportunities concerning the implementation of resources for assisting individuals suffering from mental disorders.

The comprehensive clinical benefits of digital phenotyping: from broad adoption to full impact

Digital phenotyping collects health data digitally, supporting early disease diagnosis and health management. This paper systematically reviews the diversity of research methods in digital phenotyping and its clinical benefits, while also focusing on its importance within the P4 medicine paradigm and its core role in advancing its application in biobanks. Furthermore, the paper envisions the continued clinical benefits of digital phenotyping, driven by technological innovation, global collaboration, and policy support.

Promoting mental health in the age of new digital tools: balancing challenges and opportunities of social media, chatbots, and wearables

The promotion of mental health is essential for global health, affecting millions with disorders such as anxiety and depression. Although stigma and discrimination hinder progress, these conditions are often preventable or manageable at minimal cost. The adoption of digital tools in mental health promotion, including telemedicine, online therapy, social media, and wearables, offers promising new avenues to address these issues. This review proposes a framework that focuses on the use of digital tools to enhance health literacy, foster behavioral change, and support sustained positive health behaviors. Platforms such as TikTok, Facebook, and Instagram can effectively disseminate health information, increase awareness, and enhance social accountability. Artificial intelligence-driven virtual agents offer personalised mental health interventions, providing motivational support and customised advice. Additionally, wearable technology (e.g., fitness trackers and smartwatches) enables real-time monitoring of vital health metrics, encouraging ongoing healthy activities. Nonetheless, these technologies introduce challenges including privacy issues, data security, and equitable access to digital resources, raising a new class of rights to protect mental privacy, guard against algorithm bias, and prevent personality-changing manipulations. The absence of human interaction in fully digital solutions also raises concerns about a lack of empathy and emotional connection. For optimal use of digital tools in mental health, integration with conventional care practices and adaptation to diverse cultural and social backgrounds are necessary. The results of this review suggest that digital tools, when carefully implemented, can significantly improve mental health outcomes by making care more accessible, tailored, and effective, especially for underserved communities.

Utility of Consumer-Grade Wearable Devices for Inferring Physical and Mental Health Outcomes in Severe Mental Illness: Systematic Review

BACKGROUND

Digital wearable devices, worn on or close to the body, have potential for passively detecting mental and physical health symptoms among people with severe mental illness (SMI); however, the roles of consumer-grade devices are not well understood.

OBJECTIVE

This study aims to examine the utility of data from consumer-grade, digital, wearable devices (including smartphones or wrist-worn devices) for remotely monitoring or predicting changes in mental or physical health among adults with schizophrenia or bipolar disorder. Studies were included that passively collected physiological data (including sleep duration, heart rate, sleep and wake patterns, or physical activity) for at least 3 days. Research-grade actigraphy methods and physically obtrusive devices were excluded.

METHODS

We conducted a systematic review of the following databases: Cochrane Central Register of Controlled Trials, Technology Assessment, AMED (Allied and Complementary Medicine), APA PsycINFO, Embase, MEDLINE(R), and IEEE XPlore. Searches were completed in May 2024. Results were synthesized narratively due to study heterogeneity and divided into the following phenotypes: physical activity, sleep and circadian rhythm, and heart rate.

RESULTS

Overall, 23 studies were included that reported data from 12 distinct studies, mostly using smartphones and centered on relapse prevention. Only 1 study explicitly aimed to address physical health outcomes among people with SMI. In total, data were included from over 500 participants with SMI, predominantly from high-income countries. Most commonly, papers presented physical activity data (n=18), followed by sleep and circadian rhythm data (n=14) and heart rate data (n=6). The use of smartwatches to support data collection were reported by 8 papers; the rest used only smartphones. There was some evidence that lower levels of activity, higher heart rates, and later and irregular sleep onset times were associated with psychiatric diagnoses or poorer symptoms. However, heterogeneity in devices, measures, sampling and statistical approaches complicated interpretation.

CONCLUSIONS

Consumer-grade wearables show the ability to passively detect digital markers indicative of psychiatric symptoms or mental health status among people with SMI, but few are currently using these to address physical health inequalities. The digital phenotyping field in psychiatry would benefit from moving toward agreed standards regarding data descriptions and outcome measures and ensuring that valuable temporal data provided by wearables are fully exploited.

TRIAL REGISTRATION

PROSPERO CRD42022382267; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=382267.

Development and Initial Evaluation of a Digital Phenotype Collection System for Adolescents: Proof-of-Concept Study

BACKGROUND

The growing concern on adolescent mental health calls for proactive early detection and intervention strategies. There is a recognition of the link between digital phenotypes and mental health, drawing attention to their potential use. However, the process of collecting digital phenotype data presents challenges despite its promising prospects.

OBJECTIVE

This study aims to develop and validate system concepts for collecting adolescent digital phenotypes that effectively manage inherent challenges in the process.

METHODS

In a formative investigation (N=34), we observed adolescent self-recording behaviors and conducted interviews to develop design goals. These goals were then translated into system concepts, which included planners resembling interfaces, simplified data input with tags, visual reports on behaviors and moods, and supportive ecological momentary assessment (EMA) prompts. A proof-of-concept study was conducted over 2 weeks (n=16), using tools that simulated the concepts to record daily activities and complete EMA surveys. The effectiveness of the system was evaluated through semistructured interviews, supplemented by an analysis of the frequency of records and responses.

RESULTS

The interview findings revealed overall satisfaction with the system concepts, emphasizing strong support for self-recording. Participants consistently maintained daily records throughout the study period, with no missing data. They particularly valued the recording procedures that aligned well with their self-recording goal of time management, facilitated by the interface design and simplified recording procedures. Visualizations during recording and subsequent report viewing further enhanced engagement by identifying missing data and encouraging deeper self-reflection. The average EMA compliance reached 72%, attributed to a design that faithfully reflected adolescents' lives, with surveys scheduled at convenient times and supportive messages tailored to their daily routines. The high compliance rates observed and positive feedback from participants underscore the potential of our approach in addressing the challenges of collecting digital phenotypes among adolescents.

CONCLUSIONS

Integrating observations of adolescents' recording behavior into the design process proved to be beneficial for developing an effective and highly compliant digital phenotype collection system.

Wearable Data From Subjects Playing Super Mario, Taking University Exams, or Performing Physical Exercise Help Detect Acute Mood Disorder Episodes via Self-Supervised Learning: Prospective, Exploratory, Observational Study

BACKGROUND

Personal sensing, leveraging data passively and near-continuously collected with wearables from patients in their ecological environment, is a promising paradigm to monitor mood disorders (MDs), a major determinant of the worldwide disease burden. However, collecting and annotating wearable data is resource intensive. Studies of this kind can thus typically afford to recruit only a few dozen patients. This constitutes one of the major obstacles to applying modern supervised machine learning techniques to MD detection.

OBJECTIVE

In this paper, we overcame this data bottleneck and advanced the detection of acute MD episodes from wearables' data on the back of recent advances in self-supervised learning (SSL). This approach leverages unlabeled data to learn representations during pretraining, subsequently exploited for a supervised task.

METHODS

We collected open access data sets recording with the Empatica E4 wristband spanning different, unrelated to MD monitoring, personal sensing tasks-from emotion recognition in Super Mario players to stress detection in undergraduates-and devised a preprocessing pipeline performing on-/off-body detection, sleep/wake detection, segmentation, and (optionally) feature extraction. With 161 E4-recorded subjects, we introduced E4SelfLearning, the largest-to-date open access collection, and its preprocessing pipeline. We developed a novel E4-tailored transformer (E4mer) architecture, serving as the blueprint for both SSL and fully supervised learning; we assessed whether and under which conditions self-supervised pretraining led to an improvement over fully supervised baselines (ie, the fully supervised E4mer and pre-deep learning algorithms) in detecting acute MD episodes from recording segments taken in 64 (n=32, 50%, acute, n=32, 50%, stable) patients.

RESULTS

SSL significantly outperformed fully supervised pipelines using either our novel E4mer or extreme gradient boosting (XGBoost): n=3353 (81.23%) against n=3110 (75.35%; E4mer) and n=2973 (72.02%; XGBoost) correctly classified recording segments from a total of 4128 segments. SSL performance was strongly associated with the specific surrogate task used for pretraining, as well as with unlabeled data availability.

CONCLUSIONS

We showed that SSL, a paradigm where a model is pretrained on unlabeled data with no need for human annotations before deployment on the supervised target task of interest, helps overcome the annotation bottleneck; the choice of the pretraining surrogate task and the size of unlabeled data for pretraining are key determinants of SSL success. We introduced E4mer, which can be used for SSL, and shared the E4SelfLearning collection, along with its preprocessing pipeline, which can foster and expedite future research into SSL for personal sensing.

Integration of passive sensing technology to enhance delivery of psychological interventions for mothers with depression: the StandStrong study

Psychological interventions delivered by non-specialist providers have shown mixed results for treating maternal depression. mHealth solutions hold the possibility for unobtrusive behavioural data collection to identify challenges and reinforce change in psychological interventions. We conducted a proof-of-concept study using passive sensing integrated into a depression intervention delivered by non-specialists to twenty-four adolescents and young mothers (30% 15-17 years old; 70% 18-25 years old) with infants (< 12 months old) in rural Nepal. All mothers showed a reduction in depression symptoms as measured with the Beck Depression Inventory. There were trends toward increased movement away from the house (greater distance measured through GPS data) and more time spent away from the infant (less time in proximity measured with the Bluetooth beacon) as the depression symptoms improved. There was considerable heterogeneity in these changes and other passively collected data (speech, physical activity) throughout the intervention. This proof-of-concept demonstrated that passive sensing can be feasibly used in low-resource settings and can personalize psychological interventions. Care must be taken when implementing such an approach to ensure confidentiality, data protection, and meaningful interpretation of data to enhance psychological interventions.

[Digital technologies to improve mental health]

The burden of mental diseases is enormous and constantly growing worldwide. The resulting increase in demand for psychosocial help is also having a negative impact on waiting times for psychotherapy in Germany. Digital interventions for mental health, such as interventions delivered through or with the help of a website (e.g. "telehealth"), smartphone, or tablet app-based interventions and interventions that use text messages or virtual reality, can help. This article begins with an overview of the functions and range of applications of digital technologies for mental health. The evidence for individual digital forms of interventions is addressed. Overall, it is shown that digital interventions for mental health are likely to be cost-effective compared to no therapy or a non-therapeutic control group. Newer approaches such as "digital phenotyping" are explained in the article. Finally, individual papers from the "Leibniz ScienceCampus Digital Public Health" are presented, and limitations and challenges of technologies for mental health are discussed.

Interdisciplinary perspectives on digital technologies for global mental health

Digital Mental Health Technologies (DMHTs) have the potential to close treatment gaps in settings where mental healthcare is scarce or even inaccessible. For this, DMHTs need to be affordable, evidence-based, justice-oriented, user-friendly, and embedded in a functioning digital infrastructure. This viewpoint discusses areas crucial for future developments of DMHTs. Drawing back on interdisciplinary scholarship, questions of health equity, consumer-, patient- and developer-oriented legislation, and requirements for successful implementation of technologies across the globe are discussed. Economic considerations and policy implications complement these aspects. We discuss the need for cultural adaptation specific to the context of use and point to several benefits as well as pitfalls of DMHTs for research and healthcare provision. Nonetheless, to circumvent technology-driven solutionism, the development and implementation of DMHTs require a holistic, multi-sectoral, and participatory approach.

What digital health technology types are used in mental health prevention and intervention? Review of systematic reviews for systematization of technologies

Digital health technology has been widely applied to mental health interventions worldwide. Using digital phenotyping to identify an individual's mental health status has become particularly important. However, many technologies other than digital phenotyping are expected to become more prevalent in the future. The systematization of these technologies is necessary to accurately identify trends in mental health interventions. However, no consensus on the technical classification of digital health technologies for mental health interventions has emerged. Thus, we conducted a review of systematic review articles on the application of digital health technologies in mental health while attempting to systematize the technology using the Delphi method. To identify technologies used in digital phenotyping and other digital technologies, we included 4 systematic review articles that met the inclusion criteria, and an additional 8 review articles, using a snowballing approach, were incorporated into the comprehensive review. Based on the review results, experts from various disciplines participated in the Delphi process and agreed on the following 11 technical categories for mental health interventions: heart rate estimation, exercise or physical activity, sleep estimation, contactless heart rate/pulse wave estimation, voice and emotion analysis, self-care/cognitive behavioral therapy/mindfulness, dietary management, psychological safety, communication robots, avatar/metaverse devices, and brain wave devices. The categories we defined intentionally included technologies that are expected to become widely used in the future. Therefore, we believe these 11 categories are socially implementable and useful for mental health interventions.

Reconciling healthism and techno-solutionism: An observational study of a digital mental health trial

In a growing trend in digital psychiatry, algorithmic systems are used to determine correlations between data that is collected using wearable devices and self-reports of mood. They then offer recommendations for behaviour modification for improved mood. The present study consists of observations of the development of one of these systems. Descriptions of the trial emphasise the powerful role of the intrinsically motivated, responsible participant on one hand and the empowering machine learning (ML)-based technology on the other. This conceptualisation is shown to extend the neoliberal paradox of a freedom that, to be maintained, must be continually adjusted through discipline. Because of the paradoxical nature of this formulation, laboratory members disagree about the balance of agency between the objective machine learning system and the empowered participant. The guides who help participants interpret ML outputs and implement system recommendations are ascribed a replaceable role in formal accounts. Observations of this guidance practice make clear not only the important role played by guides but also how their work is relegated to the technological side of the broader formulation of the trial and further how this conceptualisation affects the way they conduct their work.

Digital Phenotyping: Data-Driven Psychiatry to Redefine Mental Health

The term "digital phenotype" refers to the digital footprint left by patient-environment interactions. It has potential for both research and clinical applications but challenges our conception of health care by opposing 2 distinct approaches to medicine: one centered on illness with the aim of classifying and curing disease, and the other centered on patients, their personal distress, and their lived experiences. In the context of mental health and psychiatry, the potential benefits of digital phenotyping include creating new avenues for treatment and enabling patients to take control of their own well-being. However, this comes at the cost of sacrificing the fundamental human element of psychotherapy, which is crucial to addressing patients' distress. In this viewpoint paper, we discuss the advances rendered possible by digital phenotyping and highlight the risk that this technology may pose by partially excluding health care professionals from the diagnosis and therapeutic process, thereby foregoing an essential dimension of care. We conclude by setting out concrete recommendations on how to improve current digital phenotyping technology so that it can be harnessed to redefine mental health by empowering patients without alienating them.

Issues and opportunities of digital phenotyping: ecological momentary assessment and behavioral sensing in protecting the young from suicide

Digital phenotyping refers to the collection of real-time biometric and personal data on digital tools, mainly smartphones, and wearables, to measure behaviors and variables that can be used as a proxy for complex psychophysiological conditions. Digital phenotyping might be used for diagnosis, clinical assessment, predicting changes and trajectories in psychological clinical conditions, and delivering tailored interventions according to individual real-time data. Recent works pointed out the possibility of using such an approach in the field of suicide risk in high-suicide-risk patients. Among the possible targets of such interventions, adolescence might be a population of interest, since they display higher odds of committing suicide and impulsive behaviors. The present work systematizes the available evidence of the data that might be used for digital phenotyping in the field of adolescent suicide and provides insight into possible personalized approaches for monitoring and treating suicidal risk or predicting risk trajectories. Specifically, the authors first define the field of digital phenotyping and its features, secondly, they organize the available literature to gather all the digital indexes (active and passive data) that can provide reliable information on the increase in the suicidal odds, lastly, they discuss the challenges and future directions of such an approach, together with its ethical implications.