This article is more than 5 years old. Information may no longer be current.

Diagnostic accuracy of mammography not improved with computer aided-detection

Mammograms that utilize computer-aided detection did not improve cancer detection in women, according to recently published data in JAMA Internal Medicine.

“We found no evidence that [computer-aided detection] applied to digital mammography in US community practice improves screening mammography performance on any performance measure or in any subgroup of women. In fact, mammography sensitivity was decreased in the subset of radiologists who interpreted mammograms with and without [computer-aided technology]” Constance D. Lehman, MD, PhD, department of radiology at Massachusetts General Hospital, and colleagues wrote.

To assess the efficacy of digital screening mammography with and without computer-aided detection (CAD), Lehman and colleagues evaluated the accuracy of 495,818 screenings with CAD and 129,807 without CAD between 2003 and 2009. The study included 323,973 women, aged 40 to 89 years.

Results demonstrated that CAD was not associated with improved screening accuracy.

There was no difference in cancer detection rates between CAD and non-CAD mammography.

Sensitivity was comparable, with 85.3% sensitivity for mammography with CAD and 87.3% without CAD. Specificity was also comparable between CAD and non-CAD, with 91.6% and 91.4%, respectively. 

Among a subset of 107 radiologists who interpreted both CAD and non-CAD mammograms, sensitivity was significantly decreased for mammograms with CAD (OR = 0.53; 95% CI, 0.29 – 0.97).

The researchers noted that although CAD use was not associated with improved mammography interpretation, it could offer other advantages, such as improving workflow and reduced search time for faint calcifications.

“In the era of Choosing Wisely and clear commitments to support technology that brings added value to the patient experience, while aggressively reducing waste and containing costs, CAD is a technology that does not seem to warrant added compensation beyond coverage of the mammographic examination. The results of our comprehensive study lend no support for continued reimbursement for CAD as a method to increase mammography performance or improve patient outcomes,” Lehman and colleagues wrote. – by Casey Hower

Disclosures: Lehman reports receiving a grant from General Electric Healthcare (GE) and is a member of the Comparative Effectiveness Research Advisory Board for GE Healthcare. No other relevant financial disclosures were reported.