Radiology 2022
19 April 2022
Authors
Lauritzen A, Rodríguez-Ruiz von Euler-Chelpin MC, Lynge E, Vejborg I, Jensen AKG, Nielsen M, Karssemeijer N, Lillholm M
An AI-based Mammography Screening Protocol: Outcome and Radiologists Workload
Background:
Developments in artificial intelligence (AI) systems to assist radiologists in reading screening mammograms could improve breast cancer screening efficiency.
Objective:
To investigate whether Transpara could detect normal, moderate-risk, and suspicious mammograms in a screening sample to safely reduce radiologist workload and evaluate across BI-RADS densities.
Methods:
This retrospective simulation study analyzed mammographic examination data consecutively collected from January 2014 to December 2015 in the Danish Capital Region breast cancer screening program. All mammograms were scored with a Transpara score from 0 to 10, representing the risk of malignancy. During simulation, normal mammograms (score < 5) would be excluded from radiologist reading and suspicious mammograms (score > recall threshold [RT]) would be recalled. Two radiologists read the remaining mammograms. The RT was fitted using another independent cohort (same institution) by matching to the radiologist sensitivity. This protocol was further applied to each BI-RADS density. Screening outcomes were measured using the sensitivity, specificity, workload, and false-positive rate. Transpara-based screening was tested for noninferiority of sensitivity compared with radiologist screening and specificities were compared.
Results:
The study sample comprised 114421 screenings for breast cancer in 114421 women, resulting in 791 screen-detected, 327 interval, and 1473 long-term cancers and 2107 false-positive screenings. The mean age of the women was 59 years. Transpara’s sensitivity was 69.7% (779 of 1118) and was noninferior (p = .02) to the radiologist screening sensitivity of 70.8% (791 of 1118). Transpara’s specificity was 98.6% (111725 of 113 303), which was higher (p < .001) than the radiologist specificity of 98.1% (111 196 of 113303). The radiologist workload was reduced by 62.6% and 25.1% of false-positive screenings were avoided. Screening results were consistent across BI-RADS densities, although not significantly so for sensitivity.
Conclusion:
Transpara could detect normal, moderate-risk, and suspicious mammograms in a breast cancer screening program, which may reduce the radiologist workload. Transpara performed consistently across breast densities.