AI Mammography Improves Breast Cancer Detection While Reducing Radiologist Workload

A new study published in Nature Medicine reveals that artificial intelligence is transforming breast cancer screening by improving detection rates while significantly reducing the workload for radiologists.

Researchers tested an AI-supported screening approach that identifies low-risk mammograms and removes them from human review. This method allows radiologists to focus only on higher-risk cases. The study found that radiologists needed to review only about one-third of all exams, which reduced their workload by 63.6%.

At the same time, the approach improved cancer detection. The detection rate increased from 6.3 to 7.3 cases per 1,000 women. This represents a 15% improvement compared to traditional screening methods. However, the recall rate also rose slightly from 4.8% to 5.5%, meaning more patients were asked to return for additional testing. Despite this increase, the positive predictive value remained stable, which indicates that the system maintained accuracy.

The research team, led by Esperanza Elías-Cabot from Spain, analyzed data from over 31,000 women screened between 2022 and 2024. They compared two approaches: the standard double reading by radiologists and a partially automated AI-assisted workflow.

In the AI model, the system first evaluated each mammogram. It classified low-risk cases as normal, while radiologists reviewed the remaining exams with AI support. This process reduced unnecessary human involvement without compromising safety.

The study also examined different imaging technologies, including digital mammography and digital breast tomosynthesis. Both methods showed reduced workloads. However, detection improvements were more noticeable in standard mammography, while results for tomosynthesis remained stable.

These findings highlight the growing role of AI in healthcare. Hospitals can use such systems to manage increasing screening demands, especially in regions with limited medical professionals. Faster processing and improved detection could lead to earlier diagnosis and better patient outcomes.

However, the study also raises important concerns. Experts emphasize the need to carefully evaluate ethical and legal implications before fully adopting AI-driven screening. Relying on AI to exclude certain cases from human review requires strong safeguards to ensure patient safety and maintain trust.

Looking ahead, further research will focus on validating this approach across different populations and healthcare systems. Developers must also improve AI performance in advanced imaging techniques and ensure transparency in decision-making.

As healthcare systems continue to adopt digital tools, AI-supported screening could become a standard practice. This shift may not only improve efficiency but also redefine how medical professionals interact with technology in clinical settings.