International evaluation of an AI system for breast cancer screening.
McKinney, Scott Mayer
Corrado, Greg S
Kelly, Christopher J
Ledsam, Joseph R
Reicher, Joshua Jay
Young, Kenneth C
De Fauw, Jeffrey
MetadataShow full item record
McKinney, S. M., Sieniek, M., Godbole, V., Godwin, J., Antropova, N., Ashrafian, H., Back, T., et al. (2020). International evaluation of an AI system for breast cancer screening.. Nature, 577 (7788), 89-94. https://doi.org/10.1038/s41586-019-1799-6
Screening mammography aims to identify breast cancer before symptoms appear, enabling earlier therapy for more treatable disease. Despite the existence of screening programmes worldwide, interpretation of mammograms suffers from suboptimal rates of false positives and false negatives. Here we present an AI system capable of surpassing expert readers in breast cancer prediction performance. To assess its performance in the clinical setting, we curated a large representative dataset from the United Kingdom (UK) and a large enriched dataset from the United States (US). We show an absolute reduction of 5.7%/1.2% (US/UK) in false positives and 9.4%/2.7% (US/UK) in false negatives. We show evidence of the system's ability to generalise from the UK sites to the US site. In an independently-conducted reader study, the AI system out-performed all six radiologists with an area under the receiver operating characteristic curve greater than the average radiologist by an absolute margin of 11.5%. By simulating the AI system's role in the double-reading process, we maintain noninferior performance while reducing the second reader's workload by 88%. This robust assessment of the AI system paves the way for prospective clinical trials to improve the accuracy and efficiency of breast cancer screening.
Humans, Breast Neoplasms, Mammography, Reproducibility of Results, Artificial Intelligence, United States, Female, Early Detection of Cancer, United Kingdom
Professor Fiona Gilbert receives funding from the National Institute for Health Research (Senior Investigator award).
Department of Health (via National Institute for Health Research (NIHR)) (NF-SI-0515-10067)
External DOI: https://doi.org/10.1038/s41586-019-1799-6
This record's URL: https://www.repository.cam.ac.uk/handle/1810/299195
All rights reserved