New AI framework DynaBridge enhances mental health assessment using multimodal data and LLM summaries

Researchers have developed DynaBridge, a novel multimodal framework designed to dynamically fuse acoustic, visual, and textual data for mental health assessment. This framework specifically leverages Large Language Model (LLM)-generated summaries that are aware of the DASS-21 (Depression, Anxiety, and Stress Scale) structure. DynaBridge aims to improve the accuracy of predicting depression, anxiety, and stress by integrating these semantic cues with direct multimodal risk predictions and a confidence-aware refinement strategy. The model demonstrated superior performance on the AdoDAS dataset, outperforming existing multimodal methods in predicting risk categories and individual DASS-21 item responses. AI

IMPACT This research could lead to more accurate and structured AI-driven tools for mental health diagnosis and monitoring.

RANK_REASON The cluster contains a research paper detailing a new AI model and its performance on a specific task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• AdoDAS
• DASS-21
• DynaBridge

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →