AI Boosts Breast Cancer Detection and Cuts Screening Workload

Breast cancer screening programs face rising demand and persistent workforce shortages, straining double-reading workflows and delaying care. Early detection is critical to reduce mortality and minimize invasive treatment. Researchers now report that artificial intelligence (AI), evaluated within National Health Service (NHS; UK) screening services, can help address these pressures. A newly reported evaluation indicates that AI can enhance detection while reducing reading time and overall workload.

The evaluation assessed AI software developed by Google as a reader within the NHS double-reading pathway. In the United Kingdom, each screening mammogram is interpreted by two human readers. The assessment compared two human readers with one human reader plus one AI reader, and also examined AI for arbitration when human readers disagreed.

The largest NHS analysis to date encompassed 175,000 women across three study components. In a retrospective study of 125,000 women aged 50 to 70 from five NHS screening services, with 39 months of follow-up and 115,973 scans in the final analysis, using AI as the second reader increased the cancer detection rate (CDR) from 7.54 to 9.33 per 1,000 women. AI identified more invasive cancers, reduced false positives, and detected 25% of interval cancers, defined as cancers found between negative screens. For first screens, recalls fell by 39.3% and CDR rose by 8.8%.

Reading efficiency improved substantially. Across the retrospective cohort, the number of reads fell from 288,616 to 195,983, reflecting a 32.1% reduction in time spent reading. In a prospective analysis of 9,266 current cases at two screening services spanning 12 London sites, AI initially exceeded the target recall rate, prompting an adjustment to study criteria. Even after adjustment, AI maintained a higher recall rate, while completing a read in an average of 17.7 minutes versus 2.08 days for the first human reader.

A third component evaluated AI in arbitration among 50,000 women, the first such use reported. AI performed comparably to human arbitrators and, despite a higher arbitration rate, reduced overall screening workload. Investigators suggested that further development could enable earlier detection, and noted that this work will inform the UK prospective EDITH trial, described as the largest international study comparing different AI tools in mammography machines across 30 sites.

The multicenter effort involved Imperial College London, Google, the universities of Cambridge and Surrey, NHS Trusts at Cambridge University Hospitals, Imperial College Healthcare, the Royal Marsden, the Royal Surrey and St George's University Hospitals, and the AIMS public engagement group. Findings were published in two linked papers in Nature Cancer. 

“Early detection is our most powerful tool in the fight against breast cancer, and these findings mark a genuine turning point. This is the first time that we've been able to rigorously test doctors and AI working alongside each other in a clinical setting. These findings have the potential to support the transformation of the NHS, and the experiences of the people on both sides of the scan, bringing us one step closer to a future where this technology strengthens entire healthcare systems and, ultimately, saves lives,” said Dr. Susan Thomas, Clinical Director at Google and an author on both papers.

“This study provides good evidence for the potential use of AI in the real world of screening mammography, where staffing is particularly difficult. Its introduction could provide support for the successful NHS breast screening program, reducing breast cancer mortality," said Professor Deborah Cunningham, a consultant radiologist at Imperial College Healthcare NHS Trust.

"The time saved will free up radiologists to perform more hands-on tasks such as needle biopsy, an essential part of the cancer diagnostic pathway. This should not be regarded as a threat to radiologists' livelihood, rather an opportunity to allow us to spend more time deploying our skills and working with colleagues and patients to improve cancer diagnosis and outcomes,” added Professor Cunningham.