Digital Mental Health Biomarkers Dataset

Read full text Summarise with VIRAT
Authors: Juan Moisés De La Serna
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Global database of digital mental health biomarkers derived from smartphones, wearables, and passive sensing technologies (digital phenotyping). Covers clinical and observational studies (2010–2026). KEY VARIABLES: - Sleep patterns (actigraphy, wristband sensors) - Physical activity (accelerometer/steps, heart rate) - Smartphone usage (screen time, app usage, call/SMS logs) - Geolocation (GPS mobility patterns) - Digital social interaction (communication frequency) MENTAL HEALTH OUTCOMES: Depression, Anxiety, Stress, Suicide Risk KEY STATISTICS (Perplexity Research, 2024–2026): - Accelerometer, steps, HR, and sleep are core biomarkers in >80% of studies - Reduced social interactions correlate negatively with depression severity (r ≈ −0.3 to −0.5) - EDA + GPS add predictive value in >70% of subsets - Sample sizes: 30–170,320 participants (median ~189 for anxiety studies) - Study durations: 89–391 days - Multimodal models achieve AUC >0.80 for unipolar/bipolar distinction - 53% of anxiety studies focus on trait anxiety; young adults ~50% of samples - Youth studies (post-2020): aggregated features outperform single variables for depression prediction REFERENCE: Torous, J., et al. (2021). Digital phenotyping for mental health: Opportunities and challenges. World Psychiatry, 20(2), 318–335. https://doi.org/10.1002/wps.20883

Fields of Study & Subjects

Digital Mental Health Interventions Mental Health Research Topics Emotion and Mood Recognition

Publication details

Journal / source
OSF Preprints (OSF Preprints)
Publication date
23 March 2026
Language
EN
Document type
Preprint