Format

Send to

Choose Destination
Nature. 2020 Jan;577(7788):89-94. doi: 10.1038/s41586-019-1799-6. Epub 2020 Jan 1.

International evaluation of an AI system for breast cancer screening.

Author information

1
Google Health, Palo Alto, CA, USA. scottmayer@google.com.
2
Google Health, Palo Alto, CA, USA.
3
DeepMind, London, UK.
4
Department of Surgery and Cancer, Imperial College London, London, UK.
5
Institute of Global Health Innovation, Imperial College London, London, UK.
6
Cancer Research UK Imperial Centre, Imperial College London, London, UK.
7
Northwestern Medicine, Chicago, IL, USA.
8
Department of Radiology, Cambridge Biomedical Research Centre, University of Cambridge, Cambridge, UK.
9
Royal Surrey County Hospital, Guildford, UK.
10
Verily Life Sciences, South San Francisco, CA, USA.
11
Google Health, London, UK.
12
Stanford Health Care and Palo Alto Veterans Affairs, Palo Alto, CA, USA.
13
The Royal Marsden Hospital, London, UK.
14
Thirlestaine Breast Centre, Cheltenham, UK.
15
Google Health, Palo Alto, CA, USA. tsed@google.com.
16
Google Health, Palo Alto, CA, USA. sshetty@google.com.

Abstract

Screening mammography aims to identify breast cancer at earlier stages of the disease, when treatment can be more successful1. Despite the existence of screening programmes worldwide, the interpretation of mammograms is affected by high rates of false positives and false negatives2. Here we present an artificial intelligence (AI) system that is capable of surpassing human experts in breast cancer prediction. To assess its performance in the clinical setting, we curated a large representative dataset from the UK and a large enriched dataset from the USA. We show an absolute reduction of 5.7% and 1.2% (USA and UK) in false positives and 9.4% and 2.7% in false negatives. We provide evidence of the ability of the system to generalize from the UK to the USA. In an independent study of six radiologists, the AI system outperformed all of the human readers: the area under the receiver operating characteristic curve (AUC-ROC) for the AI system was greater than the AUC-ROC for the average radiologist by an absolute margin of 11.5%. We ran a simulation in which the AI system participated in the double-reading process that is used in the UK, and found that the AI system maintained non-inferior performance and reduced the workload of the second reader by 88%. This robust assessment of the AI system paves the way for clinical trials to improve the accuracy and efficiency of breast cancer screening.

PMID:
31894144
DOI:
10.1038/s41586-019-1799-6

Supplemental Content

Full text links

Icon for Nature Publishing Group
Loading ...
Support Center