<ul><li>A novel cross-modal contrastive learning framework called NeuroLIP has been proposed.</li><li>NeuroLIP integrates functional magnetic resonance imaging (fMRI) connectivity data with phenotypic textual descriptors.</li><li>The framework improves interpretability using token-level attention maps, revealing brain region-disease associations.</li><li>NeuroLIP demonstrates superiority in fairness metrics while maintaining overall best standard metric performance.</li></ul>

NeuroLIP: Interpretable and Fair Cross-Modal Alignment of fMRI and Phenotypic Text

Discover more