iCAD's AI-based 3D mammogram support tool earns FDA nod

The FDA approved iCAD’s artificial intelligence-based solution that streamlines the interpretation of breast tomosynthesis, or 3D mammograms, slashing the amount of time it takes a radiologist to read the images.

Unlike two-dimensional mammography, which generates four images of the breasts, tomosynthesis can produce hundreds of images. While tomosynthesis can provide a more complete picture, more images means a radiologist will take longer to review a patient’s case.

iCAD’s PowerLook Tomo Detection is a computer-aided detection solution for tomosynthesis. It uses a detection algorithm to pinpoint areas of interest on tomosynthesis images and combines them in a single 2D image, the company said.

“[The resulting image] not only helps reduce the reading time and improve the reading experience for radiologists, but it can also provide radiologists with an added level of confidence,” said Justin Boatsman, M.D., medical director and a diagnostic radiologist at Intrinsic Imaging, in the statement.

A U.S. study involving 20 radiologists reading 240 cases with and without PowerLook found that the technology reduced reading time by an average of 29%, with no statistically significant change in sensitivity or specificity. The tech scored a CE mark in April last year, on the back of data from six radiologists reading 80 cases that turned up similar results.

While tomosynthesis is promising, it is still relatively new. It is offered at a limited number of hospitals and not all insurance providers cover it, according to the American Cancer Society.

"As the use of 3D mammography continues to grow in the U.S. and abroad, iCAD remains committed to delivering innovative solutions that help radiologists be more efficient and confident when reading tomosynthesis exams," said iCAD CEO Ken Ferry in the statement.