The use of computer-aided detection (CAD) in screening mammography has ballooned in recent years. A decade ago, about one-third of screening mammograms were interpreted using CAD, now it’s used in the clear majority of cases.
But what impact does this tool have on interpreting digital mammograms? One of our recent top stories featured research from the Medical University of South Carolina that found a pretty limited impact with changes in diagnostic decisions after CAD was introduced. However, a closer look shows there was some improvement.
Published in the October issue of the American Journal of Roentgenology, Elodia B. Cole, MS, and colleagues’ work was a retrospective reader study evaluating CAD during the American College of Radiology Imaging Network’s Digital Mammographic Imaging Screening Trial (DMIST).
DMIST accrued women who underwent screening mammography from 2001 to 2003, and their scans were interpreted with the aid of one of a pair of commercially available CAD systems. The study looked at two groups of 14 and 15 radiologists, respectively, that each interpreted sets of 300 cases both with and without the use of CAD. Half of the cases in the study had cancer.
Results showed the average area under the curve was 0.71 without and 0.72 with the use of either CAD system. Sensitivity and specificity differences in performance were not statistically significant overall.
That’s the overall numbers, though. If readers’ interpretations were looked at individually, results showed that some did benefit significantly from the use of CAD. More than one-quarter of the radiologists in each group demonstrated sensitivity improvements.
No technology is a silver bullet, and there are concerns about the number of false-positive marks that can be produced by CAD systems. But there is potential for gains. While the overall impact in the trial from Cole and colleagues was minimal, some individuals found it improved their sensitivity. The task now is to hone CAD systems and make sure they are deployed in a way that delivers the greatest benefit.
-Evan Godt
Editor – Health Imaging
Evan joined TriMed in 2011, writing primarily for Health Imaging. Prior to diving into medical journalism, Evan worked for the Nine Network of Public Media in St. Louis. He also has worked in public relations and education. Evan studied journalism at the University of Missouri, with an emphasis on broadcast media.