Machine Learning for Depression Screening and Intervention: an Original Circadian Rhythm Score-based Methodology

2026-07-06

AI Summary

This study addresses key challenges in depression screening from large-scale behavioral data—namely, fragmented circadian rhythm metrics, limited model interpretability, and a lack of intervention guidance—by proposing a unified Circadian Rhythm Score (CRS). The CRS leverages non-negative constrained supervised representation learning to integrate multi-domain daily behaviors with near-lossless compression. An interpretable screening framework is developed using gradient-boosted trees combined with SHAP analysis, while interaction effects modeling and counterfactual regression are employed to estimate the impact of behavioral interventions. Evaluated on the CHARLS dataset (n=15,233), the approach achieves a ROC-AUC of 0.825, uncovering a nonlinear association between circadian rhythms and depression risk. The analysis further identifies actionable intervention thresholds, including an effective physical activity dose of approximately 300 MET-min/week and an optimal nap duration of about 65 minutes.

Abstract

Depression screening from large-scale behavioral data is challenged by fragmented circadian indicators, limited interpretability, and the lack of intervention-oriented analysis. Existing approaches typically analyze sleep, activity, and social behaviors in isolation, failing to capture their joint circadian structure. To address this limitation, we first propose the Circadian Rhythm Score (CRS), a composite index that compresses multi-domain daily behaviors into a unified representation of circadian rhythm. CRS is constructed to maximize discriminative power for depression screening while preserving behavioral semantics through non-negativity constraints. Empirical results demonstrate near-lossless compression, where a single CRS retains almost the full predictive capability compared with multiple raw behavioral indicators. Building upon CRS, we develop an interpretable depression screening framework based on gradient-boosted trees and SHAP analysis, revealing nonlinear and saturation-like associations between circadian rhythm and depression risk. Beyond risk prediction, we further integrate interaction modeling and counterfactual regression to estimate heterogeneous and dose-dependent behavioral effects, enabling intervention-oriented reasoning under different circadian contexts. Experiments on the China Health and Retirement Longitudinal Study (CHARLS, n=15,233), demonstrate robust screening performance (ROC-AUC=0.825) and identify actionable behavioral thresholds, including a minimum effective exercise dose of approximately 300 MET-min/week and an optimal restorative nap duration of approximately 65 minutes for sleep-deprived individuals. By bridging supervised representation learning and interpretable modeling, this work provides a scalable framework for depression screening and intervention-aware healthcare data mining.

Problem

• Depression screening
• Circadian rhythm
• Behavioral data
• Intervention-oriented analysis
• Interpretability

Innovation

• Circadian Rhythm Score
• Interpretable Machine Learning
• Depression Screening
• Counterfactual Regression
• Behavioral Intervention