Constructing the Mental Health Phenome: An Open Multimodal Dataset Linking Digital Behavior, Physical Health, and Mental Wellbeing

Shakson Isaac, Ambika Grover, Yentl Collin, John Torous, Chirag Patel

Additional Submission Instructions: For the camera-ready version, please include the author names and affiliations, funding disclosures, and acknowledgements.

Track: Track 2: Dataset Proposal Competition

Keywords: Digital phenotyping, multimodal dataset, mental health, foundation models, smartphone and wearable sensing, federated learning

Abstract: Mental health, despite its global importance, lacks a large-scale, multimodal, open dataset. We propose the Mental Health Phenome: an openly shareable collection of longitudinal smartphone behavior, wearable physiology, social media activity, and mental health measures. Enabling foundation models for predictive, interpretable, and personalized discovery across psychiatry, public health, and digital well-being. Starting with deeply phenotyped cohorts, the Mental Health Phenome is designed to scale to broader populations, combining mechanistic insights with global applicability.

Submission Number: 394