New CORE framework improves brain network learning across diverse sites

Researchers have developed a new framework called CORE to improve the analysis of brain networks from fMRI data, particularly when dealing with data from different sites. This method addresses issues where site-specific biases and averaged connectivity obscure important transient dynamics, hindering generalization. CORE works by decoupling site-specific confounders, extracting a stable population scaffold of connectivity edges, and then modeling transient pathway dynamics on this scaffold. Experiments show CORE significantly outperforms existing methods in cross-site generalization, even with variations in brain parcellation schemes. AI

IMPACT Enhances cross-site generalization for brain network analysis, potentially improving diagnostic accuracy in multi-center studies.

RANK_REASON This is a research paper detailing a new framework for analyzing fMRI data. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• ABIDE
• fMRI
• REST-meta-MDD
• SRPBS

• paper
• other

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