Tags: mit | ai | breast cancer | model

MIT AI Breast Cancer Model Said as Reliable as Radiologists

doctor examining breast image

(Oksana Krasyuk/Dreamstime.com)

By    |   Thursday, 18 October 2018 08:26 AM

An MIT artificial intelligence model is said to be as reliable as a radiologist in assessing dense breast tissue in mammograms, a study published Tuesday in the journal Radiology revealed.

There are numerous risk factors when it comes to breast cancer, but dense breast tissue alone can increase the risk of developing the disease, and Researchers at Massachusetts Institute of Technology and Massachusetts General Hospital have developed the automated model to detect it.

About 266,120 new cases of invasive breast cancer will be diagnosed in women this year alone and it is estimated that over 40 percent of women in the U.S. have dense breast tissue. Breastcancer.org noted that dense breasts can be six times more likely to develop cancer.

Compounding the issue is that dense tissue can mask cancers in mammograms, which is why 30 U.S. states mandate that women be notified if their mammograms show they have dense breast tissue.

“Breast density is an independent risk factor that drives how we communicate with women about their cancer risk,” study co-author Adam Yala said.

To address this, the team of researchers trained a deep-learning model to distinguish different types of breast tissue based on expert assessments.

When given a new mammogram, the model can then identify a density measurement that aligns closely with an expert radiologist’s opinion.

The model was set up at the breast imaging division at MGH and achieved 94 percent agreement among the hospital’s radiologists in a binary test to determine whether breasts were either heterogeneous and dense, or fatty and scattered. It matched radiologists’ assessments at 90 percent in all categories.

“It takes less than a second per image … (and it can be) easily and cheaply scaled throughout hospitals.” Yala said of the model.

Study co-author Regina Barzilay said they now aimed to “explore how to transition machine-learning algorithms developed at MIT into clinic benefiting millions of patients.”

The American Cancer Society estimates that the average risk of a woman in the U.S. developing breast cancer sometime in her life is about 12 percent.

© 2019 Newsmax. All rights reserved.

   
1Like our page
2Share
Health-News
An MIT artificial intelligence model is said to be as reliable as a radiologist in assessing dense breast tissue in mammograms, a study published Tuesday in the journal Radiology revealed.
mit, ai, breast cancer, model
342
2018-26-18
Thursday, 18 October 2018 08:26 AM
Newsmax Media, Inc.
 

Newsmax, Moneynews, Newsmax Health, and Independent. American. are registered trademarks of Newsmax Media, Inc. Newsmax TV, and Newsmax World are trademarks of Newsmax Media, Inc.

NEWSMAX.COM
America's News Page
© Newsmax Media, Inc.
All Rights Reserved