Research · MarkTechPost ·
Meet NeuroVFM: A New Neuroimaging Foundation Model Trained With Vol-JEPA on Uncurated Clinical MRI and CT Volumes
Researchers at the University of Michigan introduced NeuroVFM, a neuroimaging foundation model trained on 5.24 million uncurated clinical MRI and CT volumes. Its Vol-JEPA method extends image and video joint-embedding approaches to learn brain anatomy and pathology without radiology-report labels.