Improving breast cancer screening workflows with machine learning

2026-03-17 · Source: The latest research from Google · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Medical Imaging AI · Depth: Advanced, medium

Summary

Google Research, in partnership with NHS organizations, conducted the Artificial Intelligence in Mammography Screening (AIMS) study, publishing two companion studies in Nature Cancer to evaluate an AI-based breast cancer detection system. The first study demonstrated the AI system's standalone performance, achieving significantly higher sensitivity than the original first human reader, detecting 25% of interval cancers missed by double-reading, and successfully integrating into live NHS screening workflows with a median processing time of 17.7 minutes. The second study, a large-scale reader study, found that an AI-enabled workflow was statistically non-inferior to the traditional two-human reader workflow in overall sensitivity and specificity, while potentially reducing human reading workload by an estimated 46%. However, it also highlighted challenges, as human arbitration panels mistakenly overruled the AI's correct recall decisions on 93 positive cancer cases, emphasizing the need for improved AI explainability and trust. These studies collectively suggest that AI can enhance cancer detection and reduce workload in breast cancer screening, but require addressing operational issues like arbitration management and data drift for optimal implementation.

Key takeaway

An AI mammography system significantly boosts breast cancer detection, increasing the rate from 7.54 to 9.33 per 1,000 women and identifying 25% of interval cancers missed by human double-reads. When integrated as a second reader, it maintains non-inferior sensitivity and specificity while reducing human reading workload by 46%. This offers a scalable solution to radiologist shortages, but requires addressing human-AI trust issues (e.g., arbitration overriding correct AI recalls) and robust data drift management for safe deployment.

Topics

• Mammography AI
• Breast Cancer Detection
• Clinical Workflow Integration
• Healthcare AI
• AI System Evaluation

Best for: Computer Vision Engineer, AI Scientist, Research Scientist, Machine Learning Engineer

Related on AIssential

• How AI Ultrasound Systems Are Improving Real-Time Diagnostic Accuracy in Clinical Settings — AI ultrasound systems enhance diagnostic accuracy and expand access by automating measurements, guiding image acquisition, and accelerating interpretation.
• Medical AI gets 66% worse when you use automated labels for training, and the benchmark hides it! [R][P] — A recent study on fairness in medical segmentation for breast cancer tumors revealed that AI models perform significantly worse for younger patients, a bias attributed to the qualitative nature of the…
• This AI Model Predicts Breast Cancer Years Before Humans — Mirai uses AI to predict future breast cancer risk from mammograms, enabling personalized screening.
• Subgroup performance of a commercial digital breast tomosynthesis model for breast cancer detection — Granular subgroup analysis reveals a commercial AI model's varied performance in breast cancer detection.
• Google Research at The Check Up: from healthcare innovation to real-world care settings — Google Research showcased significant advancements in AI for healthcare at "The Check Up" event on March 17, 2026, focusing on democratizing scientific and clinical research.
• 😺 Mayo's AI spotted cancer 3 years before doctors did — AI is demonstrating superior diagnostic and triage capabilities in medicine, even with incomplete information.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The latest research from Google.