Lunit is highlighting results published March 10 in Nature Communications that validate the effectiveness of its Lunit Insight MMG in South Korea's national breast cancer screening program.
The study was led by Yun-Woo Chang, MD, from Soonchunhyang University Seoul Hospital in collaboration with breast and general radiologists from six South Korean academic hospitals. It included 24,543 women aged 40 and above who underwent routine biennial mammography as part of Korea's national breast cancer screening program between February 2021 and December 2022; it compared the performance of radiologist screening mammography interpretation with and without the use of Lunit's Insight MMG.
The study found the following:
- AI-assisted radiologists detected 13.8% more screen-detected breast cancers than those relying solely on conventional interpretation methods.
- The cancer detection rate (CDR) increased from five per 1,000 to 5.7 with AI assistance.
- Recall rates (RRs) were statistically unchanged.
- AI assistance led to a significant improvement in identifying small-sized tumors and node-negative cancers.
- AI assistance led to an improvement in CDRs for general radiologists, with a 26.4% increase from 3.87 per 1,000 to 4.89 per 1,000.
Lunit Insight MMG is being used in national breast screening programs in Australia, Sweden, Iceland, Singapore, Saudi Arabia, and Qatar, the company said.
The full study can be found here.
![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=100&q=70&w=100)






![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=112&q=70&w=112)









