AI-Powered Mammography Screening Boosts Cancer Detection in Single-Reader Settings
Posted on 12 Mar 2025
A new study has revealed that an artificial intelligence (AI)-powered solution significantly improves cancer detection in single-reader mammography settings without increasing recall rates, offering a transformative tool for countries that rely on single-reader mammography screenings.
This pioneering prospective study validated the real-world effectiveness of AI-driven mammography screening within South Korea's national breast cancer screening program. Conducted as the first large-scale, multicenter prospective study in a single-reading setting globally, the research highlights the significant improvements AI-assisted mammography interpretation brings to breast cancer detection, all while maintaining steady recall rates. Led by Soonchunhyang University Seoul Hospital (Seoul, South Korea), in collaboration with breast and general radiologists from six top academic hospitals in South Korea, the study analyzed 24,543 women aged 40 and older who underwent routine biennial mammography (2D full-field digital mammogram) between February 2021 and December 2022 as part of Korea's national screening initiative.
The study compared the performance of breast radiologists interpreting screening mammograms with and without Lunit’s (Seoul, South Korea) AI-based computer-aided detection, Lunit INSIGHT MMG. Results indicated that AI-assisted radiologists detected 13.8% more screen-detected breast cancers compared to those using traditional methods alone. The cancer detection rate (CDR) increased from 5.01 (per 1,000) to 5.70 with AI assistance, while recall rates (RRs) remained statistically unchanged, ensuring enhanced clinical effectiveness while avoiding unnecessary additional recalls. Additionally, AI assistance significantly improved the detection of smaller tumors and node-negative cancers, crucial indicators of early-stage detection. The study also assessed the impact of Lunit INSIGHT MMG on general radiologists without breast imaging expertise through a simulated retrospective study. The findings, published in Nature Communications, showed a 26.4% increase in CDRs for general radiologists, improving from 3.87 to 4.89 per 1,000, highlighting AI’s potential to assist radiologists with varying expertise.
This study builds upon Lunit INSIGHT MMG's established ability to enhance breast cancer detection. The findings are poised to influence breast cancer screening practices globally, especially in countries where single-reading mammography is standard. By demonstrating AI’s ability to improve radiologist accuracy without raising recall rates, the study provides essential data supporting AI’s role in optimizing cancer detection while alleviating radiologists’ workload. This is particularly pertinent as many countries face a shortage of specialized breast imaging professionals, making AI a crucial tool for improving efficiency and maintaining high diagnostic standards. With growing evidence supporting AI's effectiveness in breast cancer screening, Lunit is advancing its global adoption. Lunit INSIGHT MMG has already been implemented in national screening programs in Australia, Sweden, Iceland, Singapore, Saudi Arabia, and Qatar, helping healthcare systems increase efficiency and improve diagnostic accuracy.
"This latest prospective study in South Korea further solidifies Lunit INSIGHT MMG's potential as an indispensable tool in breast cancer screening. Whether in a single-reader or double-reader setting, AI can serve as a powerful force multiplier, assisting radiologists in detecting cancer earlier and more accurately," said Brandon Suh, CEO of Lunit. "As AI adoption accelerates, we remain committed to driving innovation and making AI-powered cancer screening the new standard of care. By working closely with healthcare providers worldwide, we aim to ensure that AI benefits as many patients as possible."
Related Links:
Lunit
Soonchunhyang University Seoul Hospital
