FDA clears RadNet’s breast ultrasound AI solution

The U.S. Food and Drug Administration recently cleared a breast ultrasound artificial intelligence solution from RadNet subsidiary DeepHealth, the company announced Thursday. 

The product works by automating lesion detection, characterization and reporting, helping streamline workflows for both radiologists and sonographers. RadNet said it validated the tool via a multi-reader, multi-case study involving 16 board-certified rads working at select imaging centers and hospitals. 

Following 510(k) clearance, the solution is now commercially available and eligible for reimbursement under an existing Category 3 CPT code for quantitative ultrasound tissue characterization. 

RadNet hopes to implement the product across its network of 440 centers by the end of the year. It noted that breast ultrasound is a key component of the cancer care pathway, with about 40% of women undergoing one at some point in their lives, and RadNet performing about 700,000 annually.

“It is a highly complex, operator-dependent examination, which can lead to significant variability in image acquisition, interpretation and reporting,” Jason McKellop, MD, medical director of women's imaging for RadNet California, said in a statement July 30. “With DeepHealth's breast ultrasound solution, we can achieve greater standardization of workflows, improving consistency while saving time for patients, sonographers and radiologists.”

DeepHealth said its breast imaging platform now includes applications for cancer detection in mammography and ultrasound, density assessment, breast arterial calcification classification, risk prediction and quality analytics. RadNet built out the portfolio partially through the acquisition of iCAD one year ago, for $103 million.