Using Smartphone-Based Digital Phenotyping to Predict Relapse in Serious Mental Disorders Among Slum Residents in...

Area of Science:

• Digital Health
• Mental Health Technology
• Machine Learning in Healthcare

Background:

• Serious mental illnesses (SMIs) have high relapse rates and limited continuous care access, especially in low-resource urban slums.
• Traditional clinical monitoring faces accessibility and scalability issues.
• Digital phenotyping using passive smartphone data presents a novel method to predict relapse by tracking real-world behavioral changes.

Purpose of the Study:

• To assess the feasibility and predictive accuracy of smartphone-based digital phenotyping for detecting relapse in individuals with SMIs.
• To evaluate this approach in the context of the Korail slum in Dhaka, Bangladesh.

Main Methods:

• A prospective 6-month cohort study involving 430 participants with SMIs using Android smartphones.
• Continuous passive data collection (screen time, mobility, communication frequency) via the DataDoc app.
• Monthly active data collection (symptoms, functioning) and integration with machine learning models for relapse prediction.

Main Results:

• Data collection commenced in August 2025, with 14 participants recruited as of January 2026.
• Preliminary data analysis is ongoing.
• Expected results publication in fall 2026.

Conclusions:

• This research pioneers smartphone-based digital phenotyping and machine learning for relapse prediction in LMIC slum settings.
• Findings aim to guide the development of scalable, low-cost digital interventions to bridge the mental health treatment gap.
• The study addresses critical needs in underresourced communities through innovative digital health solutions.