Reducing Radiologist Workload by Detecting Normal Mammograms with an AI System

 A. D. Lauritzen, A. Rodriguez-Ruiz, M. C. von Euler-Chelpin, E. Lynge, I. Vejborg, M. Nielsen, N. Karssemeijer, M. Lillholm

Purpose:

To investigate whether an AI system can detect normal mammographies in a breast cancer screening cohort.

Methods and materials:

This retrospective study analyzed 18020 doubly read studies from the Danish Capital Region breast cancer screening program comprised of 143 screen-detected cancers and 447 non-cancer recalls (false positives). Using the deep learning-based image analysis tool, Transpara v1.5, all studies were sorted into 10 categories based on findings from four views. A high category (10) indicates a high probability of malignancy, while a low category (1) indicates a very low probability of malignancy. Normal studies are identified as being in category 5 or less. This study examined the number of studies, and non-cancer recalls that can possibly be avoided by detecting normal studies before radiologist reading.

Results:

Using category 5 or less as threshold, 10545 (58.52%) studies were classified as normal. Included were 5 screen-detected cancers (3.5%) and 106 non-cancer recalls (23.71%). Category 1 and 2 comprised 4738 (26.29%) studies, 26 non-cancer recalls (5.82%), and 2 screen-detected cancers (1.36%) in total. Category 1 comprised 2627 (14.58%) studies, 12 non-cancer recalls (2.68%), and 0 screen-detected cancers.

Conclusion:

The results show that the AI system can successfully identify normal mammographies with very few missed screen-detected cancers. Furthermore, a substantial amount false positive studies were identified as normal. The results suggest that potentially AI systems could effectively and safely reduce the number of studies that radiologists would have to examine by a considerable amount, and several false positives could be avoided.