Artificial intelligence-based diagnosis of breast cancer by mammography microcalcification
Peer-Reviewed Publication
Updates every hour. Last Updated: 14-Aug-2025 09:11 ET (14-Aug-2025 13:11 GMT/UTC)
This study introduces a deep-learning system for rapid, automated detection and classification of tiny calcium deposits (microcalcifications) in mammograms to aid early breast cancer diagnosis. Leveraging a multi-center dataset of 4,810 biopsy-confirmed mammograms, our pipeline uses a Faster RCNN model with a feature-pyramid backbone to detect and classify microcalcifications—the pipeline requires no hand-tuned rules and provides both the overall cancer risk and highlighted lesion regions in seconds per image. On unseen test data, it achieved overall classification accuracy of 72% for discriminating between benign and malignant breasts and 78% sensitivity of malignant breast cancer prediction, marking a significant step toward AI-assisted, cost-effective breast-cancer screening that can run on standard radiology workstations.
A study conducted by researchers from the Agency for Science, Technology and Research (A*STAR), National Cancer Centre Singapore (NCCS) and National University of Singapore (NUS), in collaboration with biotech company, KYAN Technologies, has demonstrated that a precision medicine approach improves treatment selection for patients with soft tissue sarcomas (STS) in a clinical setting. Published in npj Precision Oncology in March 2025, the study’s findings support using data-driven and phenotypic screening approaches to treat STS.
Reporters note: The following briefs highlight selected cancer-related studies being presented at Digestive Disease Week® (DDW) 2025 in San Diego. You are welcome to use these items as-is, incorporate them into broader coverage, or use them as story starters for your own reporting. Reach out to Tware@thereisgroup.com to get more information or to connect with researchers for interviews.
Embargo: Each brief is embargoed until 12:01 a.m. PDT the day of its presentation, as listed at the end of each item. Alternatively, the entire package may be published after 12:01 a.m. PDT on May 5, 2025.