ScienceMarkTechPostabout 2 hours ago
NeuroVFM Foundation Model Trained on Clinical Scans
8 min read
University of Michigan researchers built NeuroVFM, a neuroimaging foundation model trained on 5.24 million clinical MRI and CT volumes from 566,915 studies. Published in Nature Medicine, the model uses Vol-JEPA to learn from uncurated clinical data, outperforming general models on brain-imaging tasks.
Level
Hype check
Tap to vote and see what everyone thinks.
