Fixing motion-corrupted medical imaging requires a more sophisticated approach than correcting a blurry JPEG. Compared to other imaging modalities like X-rays or CT scans, MRI scans provide high-quality soft tissue contrast. Unfortunately, MRI is highly sensitive to motion, with even the smallest of movements resulting in image artefacts. These artefacts put patients at risk of misdiagnoses or inappropriate treatment when critical details are obscured from the physician.
Researchers affiliated with the MIT Jameel Clinic, the epicentre of machine learning and AI at the Massachusetts Institute of Technology (MIT), have proposed a deep learning-physics approach to correct motion-corrupted MRI scans for improved patient outcomes and hospital cost-savings. The proposed technique uses an algorithm to construct a motion-free image from motion-corrupted data without changing any part of the scanning procedure. The latest study, authored by Nalini Singh, a Jameel Clinic-affiliated PhD student in the Harvard-MIT programme in health sciences and technology, was recently awarded best oral presentation at the Medical Imaging with Deep Learning conference.
According to Singh, future work could explore more sophisticated types of head motion as well as motion in other body parts. For instance, foetal MRI suffers from rapid, unpredictable motion that cannot be modelled only by simple translations and rotations.