Sign In

Communications of the ACM

ACM TechNews

Automated System Identifies Dense Tissue, a Risk Factor for Breast Cancer, in Mammograms

View as: Print Mobile App Share:
The automated deep-learning model identifies dense breast tissue in mammograms as reliably as expert radiologists.

A new automated model assesses dense breast tissue in mammograms (an independent risk factor for breast cancer) as reliably as expert radiologists.

Credit: Constance Lehman et al.

Massachusetts Institute of Technology and Massachusetts General Hospital (MGH) researchers have developed an automated system that assesses dense breast tissue in mammograms as reliably as human radiologists.

This marks the first time this kind of deep learning model has successfully been used in a clinic on real patients.

The researchers trained a deep learning model built on a convolutional neural network on a dataset of more than 58,000 randomly selected mammograms from over 39,000 women screened between 2009 and 2001. The team used about 41,000 mammograms for training and about 8,600 mammograms for testing the new system.

Each mammogram in the dataset has a standard Breast Imaging Reporting and Data System breast density rating in four categories: fatty, scattered (scattered density), heterogeneous (mostly dense), and dense.

When presented with more than 10,000 mammograms at MGH from January to May of 2018, the model achieved 94% agreement with the diagnoses of MGH radiologists.

From MIT News
View Full Article


Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account