โก Quick Summary
This scoping review highlights the potential of passive sensing and machine learning technologies for mental health monitoring using wearables and smartphones. The findings indicate promising accuracy, particularly in detecting anxiety and depression, but also reveal significant limitations that must be addressed for clinical application.
๐ Key Details
- ๐ Dataset: 42 peer-reviewed studies analyzed
- ๐งฉ Features used: Data from wearables and smartphones
- โ๏ธ Technology: Machine learning models including deep learning
- ๐ Performance: Convolutional neural networks achieved up to 92.16% accuracy in anxiety detection
๐ Key Takeaways
- ๐ Passive sensing offers a new approach to monitor mental health continuously and objectively.
- ๐ก Machine learning techniques are being applied to analyze behavioral data from wearables.
- ๐ฉโ๐ฌ Most studies focused on depression (55%) and anxiety (21%), primarily using wrist-worn devices.
- ๐ Deep learning models showed high accuracy, but traditional models remain popular for their interpretability.
- โ ๏ธ Limitations include small sample sizes and short monitoring periods, with 76% of studies having fewer than 100 participants.
- ๐ Ethical concerns were noted, with only 14% of studies addressing data anonymization.
- ๐ The study emphasizes the need for standardized protocols and larger longitudinal studies.
- ๐ฎ Future research should focus on multimodal sensor fusion and explainable AI to enhance clinical applicability.
๐ Background
Mental health issues are a growing global concern, often assessed through subjective methods that lack ecological validity. The integration of passive sensing technologies, such as wearables and smartphones, with machine learning presents an innovative solution for continuous and objective mental health monitoring. This approach could significantly enhance our understanding and management of mental health disorders.
๐๏ธ Study
This scoping review adhered to the PRISMA-ScR guidelines and systematically searched seven databases for studies published from January 2015 to February 2025. A total of 42 studies were included, focusing on the use of passive sensing technologies to monitor clinically diagnosed mental disorders, particularly depression and anxiety.
๐ Results
The review revealed that most studies employed cohort designs, with a median sample size of 60.5 participants. Key biomarkers identified included heart rate, movement index, and step count, with wrist-worn devices being the most common. Notably, deep learning models demonstrated high accuracy, achieving up to 92.16% accuracy in anxiety detection, while traditional models like random forests were favored for their interpretability.
๐ Impact and Implications
The findings from this review underscore the transformative potential of passive sensing and machine learning in mental health care. By enabling objective and continuous monitoring, these technologies could significantly improve the management of disorders such as depression and anxiety. However, addressing the identified limitations is crucial for effective clinical translation and ensuring ethical standards in data handling.
๐ฎ Conclusion
This scoping review illustrates the promising capabilities of passive sensing and machine learning in mental health monitoring. While the accuracy of these technologies is encouraging, the field must overcome methodological challenges and ethical concerns to fully realize their potential in clinical settings. Continued research and development are essential for integrating these innovations into everyday mental health care practices.
๐ฌ Your comments
What are your thoughts on the use of technology for mental health monitoring? We would love to hear your insights! ๐ฌ Share your comments below or connect with us on social media:
Passive Sensing for Mental Health Monitoring Using Machine Learning With Wearables and Smartphones: Scoping Review.
Abstract
BACKGROUND: Mental health issues have become a significant global public health challenge. Traditional assessments rely on subjective methods with limited ecological validity. Passive sensing via wearable devices and smartphones, combined with machine learning (ML), enables objective, continuous, and noninvasive mental health monitoring.
OBJECTIVE: This study aimed to provide a comprehensive review of the current state of passive sensing-based and ML technologies for mental health monitoring. We summarized the technical approaches, revealed the association patterns between behavioral features and mental disorders, and explored potential directions for future advancements.
METHODS: This scoping review adhered to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines and was prospectively registered on the Open Science Framework. We systematically searched 7 databases (Web of Science, PubMed, IEEE Xplore, Embase, PsycINFO, Scopus, and ACM Digital Library) for studies published between January 2015 and February 2025. We included 42 peer-reviewed studies that used passive sensing from wearables or smartphones with ML to monitor clinically diagnosed mental disorders, such as depression and anxiety. Data were synthesized across technical dimensions (data collection, preprocessing, feature engineering, and ML models) and clinical associations, with behavioral features categorized into 8 domains.
RESULTS: The 42 included studies were predominantly cohort designs (23/42, 55%), with a median sample size of 60.5 (IQR 54-99). Most studies focused on depression (23/42, 55%) and anxiety (9/42, 21%) using primarily wrist-worn devices (32/42, 76%) collecting heart rate (28/42, 67%), movement index (25/42, 60%), and step count (17/42, 40%) as key biomarkers. Deep learning models (eg, convolutional neural networks and long short-term memory) showed high accuracy, while traditional ML (eg, random forest) remained prevalent due to better interpretability. We identified critical limitations, including small samples (32/42, 76% with N<100), short monitoring periods (19/42, 45% <7 days), scarce external validation (1/42, 2%), and limited reporting on data anonymization (6/42, 14%).
CONCLUSIONS: While passive sensing and ML demonstrate promising accuracy (eg, convolutional neural network-long short-term memory achieving 92.16% in anxiety detection), the evidence remains constrained by three key limitations: (1) methodological heterogeneity (32/42, 76% single-device studies; 19/42, 45% with <7-day monitoring), (2) high risk of bias from small samples (median 60.5, IQR 54-99 participants) and scarce external validation (1/42, 2%), and (3) ethical gaps (only 6/42, 14% addressing anonymization). These findings underscore the technology's potential to transform mental health care through objective, continuous monitoring-particularly for depression (heart rate and step count biomarkers) and anxiety (sleep and social interaction patterns). However, clinical translation requires standardized protocols, larger longitudinal studies (โฅ3 months), and ethical frameworks for data privacy. Future work should prioritize multimodal sensor fusion and explainable artificial intelligence to bridge the gap between technical performance and clinical deployability.
Author: [‘Shen S’, ‘Qi W’, ‘Zeng J’, ‘Li S’, ‘Liu X’, ‘Zhu X’, ‘Dong C’, ‘Wang B’, ‘Shi Y’, ‘Yao J’, ‘Wang B’, ‘Lou X’, ‘Gu S’, ‘Li P’, ‘Wang J’, ‘Jiang G’, ‘Cao S’]
Journal: J Med Internet Res
Citation: Shen S, et al. Passive Sensing for Mental Health Monitoring Using Machine Learning With Wearables and Smartphones: Scoping Review. Passive Sensing for Mental Health Monitoring Using Machine Learning With Wearables and Smartphones: Scoping Review. 2025; 27:e77066. doi: 10.2196/77066