Artificial intelligence reconstructs missing data from fast MRI scans

NEW YORK, January 18, 2023 /PRNewswire/ — Artificial intelligence (AI) can reconstruct coarse-sampled, rapid magnetic resonance imaging (MRI) scans into high-quality images with similar diagnostic value to those produced by traditional MRI, according to a new study from NYU Grossman School, according to medicine and meta -AI research.

Using AI to reconstruct MRI scans, which are four times faster than standard scans, promises to expand MRI access to more patients and reduce appointment wait times, according to the study.

The study, published January 17 in radiology, is part of the fastMRI initiative founded in 2018 by NYU Langone Health and Meta AI Research (formerly Facebook). This initiative, aimed at accelerating MRI using AI, resulted in an AI model jointly developed by Meta-AI researchers and NYU Langone Imaging scientists and radiologists. It also produced the largest publicly available collection of raw MRI data, used by scientists and engineers around the world.

In a previous “proof-of-principle” study, the NYU Langone team simulated accelerated scans by removing about three-quarters of the raw data captured by traditional, slow-speed MRI scans. The fastMRI AI model then generated images that matched those created from the slower scans. In this new study, the researchers performed accelerated scans with just a quarter of the total data and used the AI ​​model to “fill in” the missing information, similar to how the brain creates images by supplementing missing visual information from the local context and previous experience. In both studies, the fastMRI scans were found to be just as accurate as traditional scans, with better quality.

“Our new study translates the results of the earlier laboratory study and applies them to actual patients,” says Dr. Michael P. Recht, Louis Marx Professor of Radiology and Chair of the Department of Radiology at NYU Grossman School of Medicine. “FastMRI has the potential to fundamentally change the way we perform MRI and make MRI more accessible to more patients.”

In the new study, a total of 170 participants received diagnostic knee MRI between January 2020 and February 2021 using a conventional MRI protocol followed by an accelerated AI protocol. Each exam was reviewed by six musculoskeletal radiologists, who looked for signs of meniscus or ligament tears and bone marrow or cartilage abnormalities. The radiologists reading the scans were not told which images were reconstructed with AI, and to limit the potential for recall bias, the readings of the standard images and the AI-accelerated images were performed at least four weeks apart.

The radiologists judged the AI-reconstructed images to be diagnostically equivalent to the conventional images for detecting cracks or anomalies and found that the overall image quality of the accelerated scans was significantly better than that of the conventional images.

“This research is an exciting step toward transferring AI-accelerated imaging into clinical practice,” said Patricia M. Johnson, PhD, assistant professor in the department of radiology at NYU Grossman School of Medicine. “It really paves the way for further innovation and advancements in the future.”

Expanding access to MRI

FastMRI, the researchers note, requires no special equipment. Technicians can program standard MRI machines to collect less data than is normally required, and the fastMRI initiative has made its data, models and code available as an open source project to other researchers and commercial MRI system manufacturers .

With fastMRI, an MRI exam that can take up to 30 minutes can be completed in less than 5 minutes, making the exam time for MRI comparable to X-rays or CT scans. Unlike these imaging modalities, however, MRI provides incredibly rich information, from showing soft tissues from different perspectives, to highlighting small cartilage abnormalities, to locating tumors in the abdomen.

“The price we traditionally pay for all this rich information is time,” said Daniel K. Sodickson, MD, PhD, chief of innovation at the Department of Radiology and director of the Center for Advanced Imaging Innovation and Research. “If we can speed up MRI to approach the speed of CT scans, we can make all this important information available to everyone.”

This study was funded by National Institutes of Health grants R01EB024532 and P41EB017183.

Media Inquiries:Allison Clair[email protected]917-301-5699

decision To download multimedia, view the original content:

SOURCE NYU Langone Health