New framework improves cross-site MDD classification using multimodal brain imaging

Researchers have developed M2LG-DG, a novel framework designed to improve the classification of major depressive disorder (MDD) using resting-state functional magnetic resonance imaging (rs-fMRI) data across different clinical sites. This multimodal approach integrates local and global brain connectivity patterns with non-imaging data, decomposing representations into shared and private components. By employing a cross-site supervised contrastive objective, M2LG-DG encourages the model to preserve diagnostic information across varying acquisition domains. The framework demonstrated strong performance, achieving an AUC of 69.48% on four held-out REST-meta-MDD sites and showing promise for other neuroimaging classification tasks. AI

IMPACT This framework could enhance the reliability of AI-driven diagnostic tools for mental health conditions across different clinical settings.

RANK_REASON Research paper detailing a new framework for medical imaging classification. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• Autism Brain Imaging Data Exchange
• M2LG-DG
• major depressive disorder
• Muhammad Asif Hasan
• REST-meta-MDD

• paper
• other

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