NYU Langone Health’s Department of Radiology is planning to release a large-scale dataset that includes more than 1.5 million MRI knee images in an ongoing effort to make MRI scans faster with AI.
The medical center announced its dataset release plans on Nov. 25 during the annual RSNA meeting in Chicago. The release is a part of the department’s Center for Advanced Imaging Innovation and Research (CAI2R) ongoing collaboration project, called fastMRI, with Facebook AI Research (FAIR). Through the project, the entities hope to share open source tools that will spur the development of AI systems to make MRI scans “10 times faster,” according to a press release.
“We hope that the release of this landmark dataset, the largest-ever collection of fully-sampled MRI raw data, will provide researchers with the tools necessary to overcome the challenges inherent in accelerating MR imaging,” Michael P. Recht, MD, chair and the Louis Marx Professor of Radiology at NYU Langone, said in a statement. “This work has the potential to not only help increase access to MR imaging, but also improve patient care worldwide.”
The initial release will include raw imaging data from nearly 1,600 cases and more than 1.5 million anonymous knee images, which were drawn from 10,000 knee MRIs and collected exclusively by NYU School of Medicine researchers. The dataset will also be fully compliant with HIPAA.
“This collaboration focuses on applying the strengths of machine learning to reconstruct high-value images in new ways. Rather than using existing images to train AI algorithms, we will radically change the way medical images are acquired in the first place,” Daniel K. Sodickson, MD, PhD, professor of radiology and neuroscience and physiology and director of CAI2R, said in a statement. “Our aim is not merely enhanced data mining with AI, but rather creating new capabilities for medical visualization to benefit human health.”
The dataset release will be the “largest public release of raw MRI data to date.” Future releases will also include data from liver and brain scans.