Lunit, a Korean medical AI imaging company, announced that its large-scale, multi-center prospective study validating the effectiveness of its AI-powered mammography analysis solution, Lunit INSIGHT MMG, for breast cancer detection has been published in Nature Communications.
Lunit's AI-powered mammography solution significantly improves breast cancer detection rates without increasing unnecessary recalls, as validated by a large-scale, multi-center prospective study published in Nature Communications.
Lunit's AI-powered mammography solution significantly improves breast cancer detection rates without increasing unnecessary recalls, as validated by a large-scale, multi-center prospective study published in Nature Communications.
The study is the "world's first" large-scale prospective study to validate the effectiveness of AI in a single reading environment, where a single radiologist interprets mammograms, as practiced in countries like Korea, according to the company.
The findings demonstrated that AI enhanced diagnostic accuracy while not increasing unnecessary recalls for additional testing.
The research, led by Professors Chang Yun-woo of the Department of Radiology at Soonchunhyang University Seoul Hospital and Ryu Jung-kyu of the Department of Radiology at Kyung Hee University Hospital at Gangdong, was conducted from February 2021 to December 2022 across six university hospitals in Korea. Other hospitals include Eulji University Nowon Eulji Medical Center, Konkuk University Medical Center, Busan Paik Hospital, and CHA Bundang Medical Center.
The study analyzed 24,543 women aged 40 and older who underwent mammography as part of Korea’s national cancer screening program.
The research team compared AI-assisted mammogram readings with traditional single readings by breast imaging specialists through one-year follow-ups.
The results showed that AI-assisted interpretation significantly increased the cancer detection rate (CDR) from 5.0 per 1,000 cases to 5.7 per 1,000 cases, marking a 13.8 percent improvement. Notably, AI did not affect the recall rate (RR), indicating that the technology improves cancer detection without leading to unnecessary additional examinations.
Further simulations demonstrated that AI-assisted readings by general radiologists increased the cancer detection rate from 3.9 to 4.9 per 1,000 cases, representing a 26.4 percent improvement. AI also proved effective in detecting early-stage breast cancers, including tumors smaller than 20 mm and those without lymph node involvement.
“This study confirms that AI can enhance cancer detection rates for both specialized breast radiologists and general radiologists in a single reading environment,” Professor Chung said. “Moreover, AI contributes to the early detection of breast cancer, improving patient outcomes.”
Lunit CEO Brandon Suh also said, “This large-scale prospective study validates the real-world impact of our AI solution in breast cancer detection.”
The results provide strong evidence supporting the adoption of Lunit’s AI technology in countries where a single radiologist interprets mammograms, such as Korea and the U.S., Suh added.
tweet
Related articles
Lunit expands AI deal with Saudi hospital operator HMG to analyze 1 million chest X-rays
Lunit appoints Craig Hadfield as Volpara Health CEO to drive growth and synergies
Lunit to present record 15 AI studies at ECR 2025 in Vienna
Lunit achieves record revenue growth driven by global expansion, innovation
Lee Han-soo corea022@docdocdoc.co.kr
See Other Articles
Copyright © KBR Unauthorized reproduction, redistribution prohibited