CAD System Offers Early Detection of Abnormalities

Dr. Rachel Brem, director of GW’s Breast Imaging Center and an advisor to the Cancer Institute for strategic initiatives, is the lead author of a recent study show-ing that a computer-aided detection system (CAD) can effectively detect breast cancers that a radiologist is more apt to initially miss.

“This technology was highly effec-tive in detecting even the smallest le-sions, with a sensitivity of 92 percent for lesions of 5 millimeters or less,” said Dr. Brem. She noted that detect-ing small lesions has long been a very challenging task for radiologists.Dr. Brem says lesion size is very important. “Cancers smaller than or equal to 1 centimeter have a better prognosis than cancers greater than a centimeter,” she said.

CAD systems assist the radiologist by processing the breast films, using algorithms to detect suspicious areas and highlighting those areas. The radiologist then re-examines the film and decides whether true areas of concern are present at the highlighted locations before making the final diagnosis.

The study, published in the March 2005 issue of the American Journal of Roentgenology, also looked at the rate of false positives—cases in which CAD highlighted areas turned out not to be cancerous.

“A large number of false-posi-tive marks can significantly hinder the usefulness of CAD by distract-ing the interpreting radiologist,” Dr. Brem said. “But our study showed 1.3 false-positive marks per image, a reasonable number that should not affect the radiologist’s interpretation, nor should it mean that more women would be called back for additional mammograms.”

“CAD is a useful tool for the improved detection of breast cancer,” said Dr. Brem. “Based on this study, CAD could reduce the number of missed cancers, as well as positively impact breast cancer prognosis, with-out increasing the patient recall rate or the physician’s workload.”