HSS to use 5.3 million exams for AI research

Register now

The Hospital for Special Surgery has completed a de-identification project for 5.3 million imaging exams to be used in research using artificial intelligence.

The New York-based facility says the project aims to improve the diagnostic accuracy of fractures and pathologies in radiology. The HSS Global Innovation Institute is working on the project with another unidentified AI company in an effort to test emerging algorithms.

The facility, which specializes in musculoskeletal health and orthopedics, contracted with Dicom Systems, Campbell, Calif., to fully de-identify the exams. Dicom is an enterprise imaging IT vendor. The de-identification process is crucial in being able to use the hospital’s archive of studies, executives say.

Also See: AI reduces time needed to generate radiation therapy plans

"This data pool is already in the hands of our AI partner, with the aim to significantly improve diagnostic accuracy of fractures and pathologies in radiology," says David King, executive director for the HSS Global Innovation Institute.

"This feature comes on the heels of our announcement to focus on the supply side of imaging data for research, clinical trials and deep learning,” says Florent Saint-Clair, executive vice president of Dicom. “Machine learning in medical imaging is a voracious process that requires massive data consumption. We're excited to serve as the on-ramp to AI for pioneering clinical and research organizations that are pushing the limits of healthcare innovation."

For reprint and licensing requests for this article, click here.