"Lunit AI Breast Cancer Diagnostic Ability Comparable to Specialists"… Comparative Study Published in International Journal

Breast Mammography AI Analysis Solution
'Lunit Insight MMG'
Published in International Journal Radiology

Medical AI company Lunit announced on the 13th that a study published in the international journal Radiology showed that the breast cancer diagnostic ability of its AI imaging analysis solution for mammography, 'Lunit Insight MMG,' is as excellent as that of specialists.

Lunit's AI-based mammography image analysis solution, 'Lunit Insight MMG'.
[Photo by Lunit]

The study was led by Professor Yan Chen's research team at the University of Nottingham, UK, and involved a total of 552 doctors, including 315 radiology specialists and 31 breast clinicians. These medical professionals are an expert group who individually analyze at least 5,000 mammography images annually.

The study applied 'PERFORMS (Personalised Breast Screening Assessment),' a standardized evaluation method used in the UK to assess the reading ability of breast screening readers.

The research team utilized 120 diagnostic cases from the UK National Breast Screening Programme (NHSBSP) that occurred between May 2018 and March 2021, comparing the individual reading results of medical personnel with those of Lunit's AI solution.

As a result of the comparison, Lunit Insight MMG demonstrated breast cancer detection ability at a level similar to that of the entire medical staff. The sensitivity, which is the accuracy in correctly diagnosing actual breast cancer cases by the 552 medical professionals, averaged 90%, and the specificity, which is the accuracy in correctly excluding non-breast cancer cases, averaged 76%. In contrast, Lunit's AI solution showed a sensitivity of 91%, which was even higher, and a specificity of 77%, similar to the medical staff's reading results.

Professor Yan Chen of the University of Nottingham explained, "In European countries, there is pressure to rapidly utilize medical AI to prepare for the shortage of specialists. To use AI in breast cancer screening, it is important to demonstrate satisfactory performance to medical personnel, and in that regard, this study provides strong evidence that AI can perform as well as specialists."

Seobum Seok, CEO of Lunit, said, "This is the first study in the UK where more than 550 doctors participated in the breast cancer screening standard test (PERFORMS) to directly compare AI and reading performance. Recently, studies verifying the technology of Lunit Insight MMG in Europe, including Sweden and the UK, have been consecutively introduced in global journals, so attention is focused on whether AI can change breast cancer diagnosis methods in Europe."

Meanwhile, Radiology, published by the Radiological Society of North America (RSNA), is an authoritative journal in the field of radiology with a global impact factor of 19.7.

This content was produced with the assistance of AI translation services.