New roadmap outlines 5 research priorities for AI in radiology

A new collaborative research roadmap was published today in Radiology, outlining five priority areas for researchers to advance AI in medical imaging.

"The scientific challenges and opportunities of AI in medical imaging are profound, but quite different from those facing AI generally,” said lead author, Curtis P. Langlotz, MD, PhD, professor of radiology and biomedical informatics at Stanford University, in a news release. “Our goal was to provide a blueprint for professional societies, funding agencies, research labs, and everyone else working in the field to accelerate research toward AI innovations that benefit patients.”

The roadmap is a product of an August 2018 workshop held at the National Institutes of Health (NIH) that was co-sponsored by the NIH, RSNA, the American College of Radiology and The Academy for Radiology and Biomedical Imaging Research.

In it are five research priorities:

1. Novel image reconstruction techniques that quickly produce images humans can read from source data.

2. A focus on automated image labeling and annotation, which includes “information extraction from the imaging report, electronic phenotyping and prospective structure image reporting.”

3. Machine learning models for clinical data, including pre-trained and distributed learning techniques.

4. Algorithms capable of explaining their findings to users.

5. Methods for deidentifying images and sharing image datasets that are adequately validated.

“Most imaging research laboratories are now employing machine learning methods to solve computer vision problems. Yet machine learning research is still in its early stages,” Langlotz and colleagues wrote. “Standards bodies, professional societies, government agencies, and private industry must work together to accomplish these goals in service of patients, who are sure to benefit from the innovative imaging technologies that will result.”