MIT researchers combine deep learning and physics to fix motion-corrupted MRI scans

Researchers at MIT have developed a deep learning model capable of correcting motion artifacts in brain MRI scans, improving diagnostic accuracy and potentially reducing healthcare costs associated with repeated imaging sessions.