CoDaS: AI Co-Data-Scientist for Biomarker Discovery via Wearable Sensors

TL;DR

CoDaS is an AI multi-agent system that systematically discovers and validates digital biomarkers from wearable sensor data for mental health and metabolic conditions, improving predictive models.

Contribution

Introduces CoDaS, a novel multi-agent framework that automates hypothesis generation, validation, and biomarker discovery from large-scale wearable datasets with human oversight.

Findings

Identified 41 digital biomarkers for mental health and 25 for metabolic outcomes.

Validated circadian instability features in depression cohorts with significant correlations.

Improved predictive performance for depression and insulin resistance using CoDaS-derived features.

Abstract

Scientific discovery in digital health requires converting continuous physiological signals from wearable devices into clinically actionable biomarkers. We introduce CoDaS (AI Co-Data-Scientist), a multi-agent system that structures biomarker discovery as an iterative process combining hypothesis generation, statistical analysis, adversarial validation, and literature-grounded reasoning with human oversight using large-scale wearable datasets. Across three cohorts totaling 9,279 participant-observations, CoDaS identified 41 candidate digital biomarkers for mental health and 25 for metabolic outcomes, each subjected to an internal validation battery spanning replication, stability, robustness, and discriminative power. Across two independent depression cohorts, CoDaS surfaced circadian instability-related features in both datasets, reflected in sleep duration variability (DWB, \rho = 0.252, p < 0.001) and sleep onset variability (GLOBEM, \rho = 0.126, p < 0.001). In a metabolic cohort, CoDaS derived a cardiovascular fitness index (steps/resting heart rate; \rho = -0.374, p < 0.001), and recovered established clinical associations, including the hepatic function ratio (AST/ALT; \rho = -0.375, p < 0.001), a known correlate of insulin resistance. Incorporating CoDaS-derived features alongside demographic variables led to modest but consistent improvements in predictive performance, with cross-validated \Delta R^2 increases of 0.040 for depression and 0.021 for insulin resistance. These findings suggest that CoDaS enables systematic and traceable hypothesis generation and prioritization for biomarker discovery from large-scale wearable data.