“This article may be correct, but the method of the investigation is not right. To prove that “computer-aided detection of breast cancer in mammograms” is not more accurate than with using the computer detection requires a lot more work than was done here. The proof that will stand the tests of use over time is by proper research in the lab and radiology suites by dedicated experts. Reviewing the old records of 324,000 women, called researching by meta-analysis, is suspect. Computers are drawing conclusions because they can combine large numbers of different medical reports. Those findings should not be considered proof, just interesting.” Bill Chesnut, MD
Mammograms with computer-aided detection may not be more accurate than those without.
The AP (9/29, Tanner) reports that research suggests that “computer-assisted detection used in most U.S. mammograms adds no benefit to breast cancer screening while substantially increasing costs.” The study was published in JAMA Internal Medicine. The research “involved nearly 324,000 women who had digital mammograms from 2003 to 2009.”
TIME (9/29, Sifferlin) reports that “there were 495,818 mammograms with CAD and 129,807 without and the results were interpreted by 271 radiologists from 66 different facilities.” The investigators “concluded that CAD did not improve diagnostic accuracy and overall there was no beneficial impact of CAD on mammography interpretation.”