CT breast density reading may improve risk assessment

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
 - anxious woman

Analyzing breast density at CT of the chest could offer additional information on a patient’s risk assessment for breast cancer, according to a study published online Oct. 14 in Radiology.

Mary Salvatore, MD, of the Mount Sinai School of Medicine in New York City, and colleagues performed a pilot study comparing radiologists’ reading of breast density at chest CT with breast density readings from mammography done for the same patient. A subset of these readings were then compared with computer-derived measurements of breast density at CT.

The researchers selected two radiologists with expert experience in interpreting mammographic and CT findings to independently perform a retrospective review of mammograms and chest CT scans from 206 women aged 29-91. The radiologists were asked to classify each case into one of the four breast density types defined by the Breast Imaging Reporting and Data System (BI-RADS) of the American College of Radiology.

Computer-derived measurements were then used to identify the approximate breast region of interest (ROI). The computer-based breast density was determined by the percentage of glandular tissue in the breast ROI and classified into one of the four CT density grades. Statistical analysis was completed to find agreement between the breast density type on the basis of the mammogram and chest CT scan for each radiologist. Intrareader agreement was calculated, as well as the percentage agreement between the computer-derived measurements and the consensus-read CT scans.

Results indicated that interreader agreement was higher for the CT density grades (0.79) than for the mammographic density types (0.62), which demonstrates that the CT agreement was significantly better than the mammographic agreement. The intrareader reliability of breast density grades on CT images was 0.88. The computer-derived breast density measurements agreed with those of the radiologists in 90 percent of the cases. When the four cases were manually adjusted for complex anatomy, agreement was found for all cases.

“CT breast density readings represent an opportunity to provide additional information about the risk of breast cancer that is readily available and currently not being used in a standardized manner,” wrote Salvatore and colleagues.