Key clinical point: An artificial intelligence computer algorithm performed on par with, and in some cases exceeded, radiologists in reading mammograms from women undergoing routine screening.
Major finding: When operating at a specificity of 96.6%, the sensitivity was 81.9% for the algorithm, 77.4% for first-reader radiologists, and 80.1% for second-reader radiologists.
Study details: A comparison of algorithm and radiologist assessments of mammograms in 8,805 women, 739 of whom were diagnosed with breast cancer.
Disclosures: The research was funded by the Stockholm County Council. The investigators disclosed financial relationships with the Swedish Research Council, the Swedish Cancer Society, Stockholm City Council, Collective Minds Radiology, and Pfizer.
Salim M et al. JAMA Oncol. 2020 Aug 27. doi: 10.1001/jamaoncol.2020.3321.