MoodAngels: A Retrieval-augmented Multi-agent Framework for Psychiatry Diagnosis
Mengxi Xiao, Ben Liu, He Li, Jimin Huang, Qianqian Xie, Xiaofen Zong, Mang Ye, Min Peng
TL;DR
MoodAngels tackles mood-disorder diagnosis under data scarcity and symptom overlap by introducing a retrieval-augmented multi-agent framework that performs granular item-level analysis and multi-step verification. It combines DSM-5 knowledge with anonymized historical cases in a retrieval datastore and synthesizes independent agent opinions through structured debates among three diagnostic variants (Angel.R, Angel.D, Angel.C). The open-source MoodSyn dataset provides privacy-preserving, clinically plausible synthetic data for research, while real-world experiments show MoodAngels surpasses GPT-4o by 12.3% in accuracy with further gains from the multi-agent setup; ablation studies highlight the value of granular analysis and careful scale selection. Together, these contributions advance AI-assisted psychiatric diagnosis and offer a practical research resource for computational psychiatry with strong privacy safeguards.
Abstract
The application of AI in psychiatric diagnosis faces significant challenges, including the subjective nature of mental health assessments, symptom overlap across disorders, and privacy constraints limiting data availability. To address these issues, we present MoodAngels, the first specialized multi-agent framework for mood disorder diagnosis. Our approach combines granular-scale analysis of clinical assessments with a structured verification process, enabling more accurate interpretation of complex psychiatric data. Complementing this framework, we introduce MoodSyn, an open-source dataset of 1,173 synthetic psychiatric cases that preserves clinical validity while ensuring patient privacy. Experimental results demonstrate that MoodAngels outperforms conventional methods, with our baseline agent achieving 12.3% higher accuracy than GPT-4o on real-world cases, and our full multi-agent system delivering further improvements. Evaluation in the MoodSyn dataset demonstrates exceptional fidelity, accurately reproducing both the core statistical patterns and complex relationships present in the original data while maintaining strong utility for machine learning applications. Together, these contributions provide both an advanced diagnostic tool and a critical research resource for computational psychiatry, bridging important gaps in AI-assisted mental health assessment.