Development of prediction models for screening depression and anxiety using smartphone and wearable-based digital phenotyping: protocol for the Smartphone and Wearable Assessment for Real-Time Screening of Depression and Anxiety (SWARTS-DA) observational study in Korea

Introduction Depression and anxiety are highly prevalent and burdensome, yet detection rates remain low due to limitations of current screening methods. With widespread adoption of smartphones and consumer wearables, digital phenotyping enables unobtrusive screening in natural settings. The SWARTS-DA study aims to develop prediction algorithms to identify individuals at risk for depressive and anxiety disorders and classify mild-to-severe symptom levels, by collecting comprehensive smartphone and smartwatch data to translate AI-based early detection into clinical impact. Methods and analysis This cross-sectional observational study will enroll up to 2500 participants (minimum 1000) aged 19–59 years in South Korea via social media outreach and clinical referrals. Each participant will be followed for 4 weeks, using a custom smartphone app (PixelMood). Active data include daily, weekly and monthly self-report measures (eg, PHQ-9, GAD-7), while passive smartphone data include physical activity, geolocation, ambient light and usage patterns. Optionally, Apple Watch or Galaxy Watch users will contribute additional physiological and sleep health data. The primary outcome is the development of machine-learning algorithms (eg, random forest, support vector machine, deep learning) to predict depression and anxiety using digital biomarkers. Secondary outcomes include associations between digital biomarkers and clinical measures, and feasibility/acceptability of data collection. Features reflecting mobile usage, physical/social activity, sleep patterns, resting physiology and circadian rhythms will be engineered for multimodal modeling. Ethics and dissemination Approved by Korea University Anam Hospital IRB (2023AN0506). Results will be disseminated at conferences and via peer-reviewed publications. Trial registration CRIS KCT0009183.

Yu-Bin Shin, Ah Young Kim, Seonmin Kim, Min-Sup Shin, Jinhwa Choi, Kyung Lyun Lee, Jisu Lee, Sangwon Byun, Sujin Kim, Heon-Jeong Lee, Chul-Hyun Cho