Lunit Achieves Milestone at ECR 2025: 13 Oral Presentations Among 15 AI Studies, Demonstrating Unmatched Clinical Impact

• Lunit to present 15 studies, with an unprecedented 13 accepted as oral presentations, marking a significant milestone in AI-powered radiology research
• Key studies demonstrate AI's capability to replace one radiologist in double reading, detect subclinical breast cancer years in advance, and maintain robust performance despite image quality challenges

SEOUL, South Korea, Feb. 27, 2025 /PRNewswire/ -- Lunit (KRX:328130.KQ), a leading provider of AI-powered solutions for cancer diagnostics and therapeutics, today announced a landmark achievement at the European Congress of Radiology (ECR) 2025, taking place in Vienna, Austria, from February 26 to March 2. With 15 research abstracts accepted and 13 designated as oral presentations, this year's participation marks Lunit's largest and most influential presence at ECR to date. 

The high acceptance rate underscores Lunit's growing leadership in AI-driven radiology, with its research recognized as highly relevant to the future of cancer screening and detection. Among the featured studies, two highlight groundbreaking findings that further solidify Lunit INSIGHT MMG as a clinically proven tool in mammography interpretation and workflow optimization. 

Replacing One Radiologist with AI for Independent Double Reading in Mammographic Screening

Presented by Marie Burns Bergan from BreastScreen Norway, this study evaluates the impact of replacing one radiologist with Lunit INSIGHT MMG in a large-scale double-reading mammography screening program. The research analyzed over 1 million mammography screenings, testing various AI thresholds to assess AI's role in maintaining detection accuracy while reducing radiologists' workload. 

Key findings include:

• AI successfully detected up to 79.9% of screen-detected cancers at a 10% positivity threshold.
• At a 5% threshold, AI maintained 75.5% cancer detection while also identifying 5.7% of interval cancers, which are often missed by traditional screening.
• AI integration reduced radiologists' reading workload by 50%, optimizing resources without significantly compromising detection accuracy. 

This study highlights Lunit INSIGHT MMG's ability to serve as a reliable second reader, supporting AI's role in improving efficiency and alleviating radiologist shortages in high-volume screening programs. 

Using Artificial Intelligence to Detect Subclinical Breast Cancer

Led by Jonas Gjesvik from BreastScreen Norway, this study investigates whether Lunit INSIGHT MMG can identify subtle cancer risk indicators years before a formal diagnosis. The analysis involved 116,000 women who participated in a long-term mammography screening program, assessing AI's ability to assign risk scores and track disease progression. 

Key findings include: 

• AI assigned higher risk scores to breasts that later developed cancer, even up to six years before diagnosis.
• AI scores for screen-detected breast cancers increased significantly from 19.2 to 82.7 across consecutive screenings, indicating early disease detection. 
• For interval cancers, AI identified elevated risk earlier than traditional methods, paving the way for more personalized screening strategies. 

The study underscores AI's potential to enhance breast cancer risk prediction, shifting screening protocols toward a personalized, risk-based model. This research aligns with ongoing work leveraging Volpara's breast density analysis, further strengthening the Lunit-Volpara ecosystem in AI-driven breast cancer risk assessment. 

In addition to these highlighted studies, Lunit will present a total of 15 research abstracts, further demonstrating AI's role in enhancing diagnostic accuracy, optimizing workflow, and supporting radiologists worldwide. 

"This year's ECR is a landmark moment for Lunit, with our highest-ever number of oral presentations, underscoring our leadership in AI-powered radiology research," said Brandon Suh, CEO of Lunit. "From replacing a radiologist in double-reading settings to predicting breast cancer years in advance, these findings reinforce AI's potential to transform breast cancer detection, optimize radiology workflows, and ultimately improve patient outcomes." 

Join Lunit at booth AI-19 in the EXPO X1 hall to explore the latest advancements in AI-powered cancer detection and discuss potential collaborations in the evolving field of AI-driven radiology. 

Presentations featuring Lunit INSIGHT at ECR 2025

• Lunit INSIGHT MMG, Oral Presentation, RPS-2202 "Associations of automatically measured breast density with breast cancer risk and duration of the pre-clinical detectable phase in a Dutch screening cohort" | March 2 at 08:00-09:00 CET, Research Stage 1
• Lunit INSIGHT MMG, Oral Presentation, RPS-2202 "Impact of Breast Density Metrics on Personalized Breast Cancer Screening Protocols" | March 2 at 08:00-09:00 CET, Research Stage 1
• Lunit INSIGHT MMG, Oral Presentation, RPS-2202 "Changes in Mammographic Density for Breasts Developing and not Developing Breast Cancer" | March 2 at 08:00-09:00 CET, Research Stage 1
• Lunit INSIGHT MMT, EPOS, C-16155 "Breast volume for women with screen-detected and interval cancer in BreastScreen Norway" 
• Lunit INSIGHT MMG, Oral Presentation, RPS-1002 "Replacing one radiologist with AI for independent double reading in mammographic screening" | Feb 27 at 14:00-15:30 CET, Research Stage 4
• Lunit INSIGHT MMG, Oral Presentation, RPS-1002 "The application of artificial intelligence to enhance the identification of previously missed non-palpable breast carcinomas" | Feb 27 at 14:00-15:30 CET, Research Stage 4
• Lunit INSIGHT MMG, Poster Presentation, C-18295 "Radiologist's interpretation scores on cancer cases among the screening examinations with the highest AI malignancy scores in BreastScreen Norway"
• Lunit INSIGHT MMG, Oral Presentation, RPS-2202 "Using Artificial Intelligence to Detect Subclinical Breast Cancer" | March 2 at 08:00-09:00 CET, Research Stage 1
• Lunit INSIGHT MMG, Oral Presentation, RPS-1002 "Mammographic features of false positive AI markings on screening mammograms from BreastScreen Norway" | Feb 27 at 14:00-15:30 CET, Research Stage 4
• Lunit INSIGHT MMG, Oral Presentation, RPS-705 "Artificial intelligence mammography interpretation systems are affected more by mammographic image quality issues than radiologists are" | Feb 27 at 08:00-09:30 CET, Research Stage 3
• Lunit INSIGHT MMG, Oral Presentation, RPS-705 "Repurposed AI-Based Mammography Interpretation in Diverse Clinical Scenarios" | Feb 27 at 08:00-09:30 CET, Research Stage 3
• Lunit INSIGHT MMG, Oral Presentation, RPS-1902 "Artificial intelligence as an initial reader for double reading in breast cancer screening: A prospective initial study of 32,822 mammograms of the Egyptian population" | March 1 at 12:30-13:30 CET, Research Stage 1
• Lunit INSIGHT MMG, Oral Presentation, S10 - Students session 1 "Menopausal status and breast density" | Feb 27 at 14:00-15:30 CET, Room M1
• Lunit INSIGHT DBT, Oral Poster Presentation, C-12965 "Acceptability of tomosynthesis in breast cancer screening to participants and screening professionals in the Netherlands" | Feb 26 at 15:30 CET, Poster Presentation Stage B
• Lunit INSIGHT MMG, Oral Presentation, RPS-1705 "Exploring how AI influences human gaze behaviour during mammography reading" | March 1 at 08:00-09:00 CET, Research Stage 4

About Lunit

Founded in 2013, Lunit (KRX:328130.KQ) is a medical AI company on a mission to conquer cancer. Lunit harnesses AI-powered medical image analytics and biomarker analysis to ensure accurate diagnosis and optimal treatment for each cancer patient. The FDA-cleared Lunit INSIGHT suite for cancer screening serves over 4,800 hospitals and medical institutions across 55+ countries. Lunit clinical studies have been published in top journals, including the Journal of Clinical Oncology and the Lancet Digital Health, and presented at global conferences such as the ASCO and RSNA. Headquartered in Seoul, South Korea, with a network of offices worldwide, Lunit leads the global fight against cancer. Discover more at lunit.io.

SOURCE Lunit