Meet NeuroVFM: A New Neuroimaging Foundation Model Trained With Vol-JEPA on Uncurated Clinical MRI and CT Volumes

Researchers at the University of Michigan have developed NeuroVFM, a foundation model trained on over 5 million uncurated clinical MRI and CT scans. By utilizing the Vol-JEPA architecture, the model learns to identify brain anatomy and pathologies without requiring manual labels from radiologists. This approach aims to improve the efficiency of medical image analysis by automating feature extraction from large datasets of volumetric clinical data.