AI revolutionises breast cancer screening: Study supports AI as human reader alternative

Lunit, a leader in AI-powered solutions for cancer diagnostics, conducted a collaborative study led by Dr. Fredrik Strand at Sweden's Karolinska Institutet involving over 50,000 women, supporting the use of AI as a viable alternative to one human reader in Europe's double reading system for breast cancer screening. 

The study utilised Lunit INSIGHT MMG, an FDA-cleared AI solution for mammography analysis, and found that AI, in collaboration with one radiologist, exhibited a superior Cancer Detection Rate (CDR) and significantly reduced recall rates (RR) compared to the traditional two-radiologist approach. 

Lunit has partnered with Capio S:t Göran Hospital, Sweden's largest private hospital, to implement AI as an independent reader in breast cancer screening, aiming to enhance efficiency and patient care. 

The study results were published in The Lancet Digital Health and will be presented at the EUSOBI 2023 conference in Spain. This marks a significant milestone in AI's role in redefining breast cancer screening standards and addressing radiologist shortages.