Computer-Aided Detection Reduces Mammogram Accuracy

Implementation of computer-aided detection decreases diagnostic specificity, increases biopsy rates

WEDNESDAY, April 4 (HealthDay News) -- The use of computer-aided detection systems for screening mammography has reduced the accuracy of mammogram interpretation, researchers report in the April 5 issue of the New England Journal of Medicine.

Joshua Fenton, M.D., from the University of California Davis in Sacramento, and colleagues examined data from 222,135 women (429,345 mammograms), including 2,351 women diagnosed with breast cancer within one year of the screening. The researchers measured the sensitivity, specificity and positive predictive value of computer-aided detection. The screening mammograms were conducted between 1998 and 2002 at 43 facilities, including seven that used computer-aided detection.

Use of computer-aided detection resulted in a drop in diagnostic specificity from 90.2 to 87.2 percent, a decrease in positive predictive value from 4.1 to 3.2 percent, and an increase in the rate of biopsy by 19.7 percent. The rate of cancer detection was unchanged and, overall, the accuracy was lower for computer-aided detection than for non-use.

Screening mammography is fraught with a number of difficulties including a lack of interpretive expertise, long training time and low sensitivity for certain breast and tumor types, according to an editorial by Ferris M. Hall, M.D., of Beth Israel Deaconess Medical Center in Boston. He suggests that genetic testing in combination with magnetic resonance imaging may provide more sensitive screening results.

The study authors report affiliations with R2 Technologies, Koning, Biolucent and GE Medical Systems.

Full Text (subscription or payment may be required)

Physician's Briefing