RSNA 2016: Masking index shows promise in predicting the probability of masking in a mammogram

This year’s annual RSNA conference featured focus on "Hot Topics in Breast Imaging," which outlined a variety of sub-topics including over diagnosis, trends in breast density assessment over time and breast cancer screening.

James G. Mainprize, PhD, and colleagues outlined his findings in the presentation, “A Masking Index to Predict Reduced Sensitivity of Mammography Due to Breast Density." Mainprize found that a quantitative, objective measure of masking could become a key tool in determining when the performance of screening mammography in an individual woman is compromised due to breast density.

“We are refining a masking index based on the local 'detectability map' that predicts the probability of missing a cancer (if present) due to the amount and patterns of dense tissue in the breast. The maps are automatically calculated from the image and the DICOM header contents,” noted Mainprize et al. 

Simulated breast cancer lesions were sequentially inserted into digital mammograms one by one. To create a detectability map in which pixels show the probability of detecting possible cancer in each area in the mammogram, an automated computer observer was used to combine measured image features.

“From the map, a masking index, giving the probability of missing a cancer due to reduced contrast or clutter caused by superposition of dense breast tissue, is calculated for the entire mammogram,” noted Mainprize.

The researchers studied de-identified mammograms consisting of 8 cancers missed on mammography and 40 screen-detected cancers. A larger set of 106 interval cancers and 596 screen-detected cancers are currently undergoing study as well.

Data suggests that detectability is a good classifier for masking, while Breast Imaging Reporting and Data System density assessment alone has limitations and was not as informative in predicting missed cancer status. 

“A masking index is being developed that has shown promise in predicting the probability of masking in a mammogram.  It is based on detectability maps that indicated regions of high/low detectability and have been shown to correlate with radiologists’ impressions of mammograms and screening performance," he said. 

This information could serve as an important tool for future determinations on when the performance of screening mammography in an individual woman is compromised due to breast density.