All Sections

Home | Business

Software helping radiologists quickly detect breast cancer gets recognition

Union Leader Correspondent

October 15. 2017 9:12PM
Diane Clifford, director of marketing for iCAD Inc., explains the company's PowerLook Tomo Detection product during the New Hampshire High Tech Council's 12th annual Product of the Year award held at Manchester Country Club in Bedford on Thursday. (THOMAS ROY/UNION LEADER)

PowerLook Tomo Detection was developed by a team of Nashua research scientists and developers at iCAD Inc. The software helps radiologists quickly detect breast cancer. (COURTESY)

NASHUA — A team of local research scientists has devoted the past three years to developing a new product that helps detect breast cancer.

Last week, that team at iCAD Inc. of Nashua was awarded and recognized for its PowerLook Tomo Detection software, which received the 2017 Product of the Year award from the New Hampshire High Technology Council.

“This was a team effort. There were lots of researchers and other people involved,” said Senthil Periaswamy, the lead scientist who oversaw the research and development for the medical imaging product.

Periaswamy, 48, who has worked at iCAD for the past decade, said his company was committed to developing a product that would help radiologists scan through hundreds of images quickly and allow them to accurately detect breast cancer using three dimensional mammography or tomosynthesis technology.

“(Radiologists) are human beings, and they are having more fatigue when they work with these large amounts of images,” Periaswamy said. “They are human beings and can miss things.”

The PowerLook Tomo Detection, which is now approved by the Food and Drug Administration and is already being sold on the market, is being described as a necessary spell-check for radiologists.

The software provides an algorithm that is trained using examples of different types of cancers. The product is shown the varieties of cancers, and when a new case is scanned, the software tries to learn from the examples and find new cancer in the images, the developers say.

“We use a lot of data — more than 5,000 cases to train with,” said Periaswamy, who said the software allowed iCAD researchers to use deep learning technology to boost performance.

Two-dimensional digital mammography typically produces four images per exam, while digital breast 3-D tomosynthesis can produce hundreds of images, which significantly increases the time it takes for radiologists to interpret exams.

PowerLook Tomo Detection improves radiologists’ efficiency by automatically analyzing each tomosynthesis plan and identifying suspicious areas. Those areas are then blended onto a 2-D synthetic image to provide radiologists with a single enhanced image that is used to more efficiently navigate the large tomosynthesis data set, the company says.

“As the use of 3-D 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,” Ken Ferry, chief executive officer of iCAD, said in an earlier statement.

The product was developed in Nashua at iCAD’s facility at 98 Spit Brook Road and is available on GE Healthcare digital breast tomosynthesis systems.

Business Nashua

More Headlines

Motivation Matters: A noble use of power