Purpose of Review To review research on deep learning models and their potential application within breast screening. Recent Findings The greatest issue in breast screening is a workforce crisis across the UK, much of Europe and even Japan. Traditional computer-aided detection (CAD) for mammography decision-support could not reach the level of an independent reader. Deep learning (DL) outperforms CAD and is close to surpassing human performance. DL is already capable of decision support and density assessment for 2D full-field digital mammography (FFDM), and is on the cusp of providing consistent, accurate and interpretable mammography reading as an independent reader. Summary A bold vision for the future of breast cancer screening is required if programmes are to maintain double reading standards. DL provides the potential for single reading programmes, such as in the USA, to reach EU double reading accuracy, as well as providing practical support for adoption of the emerging modality of digital breast tomosynthesis.

No abstract

No abstract

Objective To evaluate the effectiveness of a new strategy for using artificial intelligence (AI) as supporting reader for the detection of breast cancer in mammography-based double reading screening practice. Methods Large-scale multi-site, multi-vendor data were used to retrospectively evaluate a new paradigm of AI-supported reading. Here, the AI served as the second reader only if it agrees with the recall/no-recall decision of the first human reader. Otherwise, a second human reader made an assessment followed by the standard clinical workflow. The data included 280 594 cases from 180 542 female participants screened for breast cancer at seven screening sites in two countries and using equipment from four hardware vendors. The statistical analysis included non-inferiority and superiority testing of cancer screening performance and evaluation of the reduction in workload, measured as arbitration rate and number of cases requiring second human reading. Results AI as a supporting reader was found to be superior or non-inferior on all screening metrics compared with human double reading while reducing the number of cases requiring second human reading by up to 87% (245 395/280 594). Compared with AI as an independent reader, the number of cases referred to arbitration was reduced from 13% (35 199/280 594) to 2% (5056/280 594). Conclusion The simulation indicates that the proposed workflow retains screening performance of human double reading while substantially reducing the workload. Further research should study the impact on the second human reader because they would only assess cases in which the AI prediction and first human reader disagree.

Objectives To evaluate the effectiveness of a novel strategy for using AI as a supporting reader for the detection of breast cancer in mammography-based double reading screening practice. Instead of replacing a human reader, here AI serves as the second reader only if it agrees with the recall/no-recall decision of the first human reader. Otherwise, a second human reader makes an assessment, enacting standard human double reading. Design Retrospective large-scale, multi-site, multi-device, evaluation study. Participants 280,594 cases from 180,542 female participants who were screened for breast cancer with digital mammography between 2009 and 2019 at seven screening sites in two countries (UK and Hungary). Main outcome measures Primary outcome measures were cancer detection rate, recall rate, sensitivity, specificity, and positive predictive value. Secondary outcome was reduction in workload measured as arbitration rate and number of cases requiring second human reading. Results The novel workflow was found to be superior or non-inferior on all screening metrics, almost halving arbitration and reducing the number of cases requiring second human reading by up to 87.50% compared to human double reading. Conclusions AI as a supporting reader adds a safety net in case of AI discordance compared to alternative workflows where AI replaces the second human reader. In the simulation using large-scale historical data, the proposed workflow retains screening performance of the standard of care of human double reading while drastically reducing the workload. Further research should study the impact of the change in case mix for the second human reader as they would only assess cases where the AI and first human reader disagree.