A major Swedish clinical trial has found that using artificial intelligence (AI) to assist with routine mammography screening leads to earlier breast cancer detection and a significant reduction in cancers missed between screenings. The results, published January 30 in The Lancet Digital Health, come from the MASAI trial—a first-of-its-kind, large-scale randomized study involving over 105,000 women screened between April 2021 and December 2022.
AI Boosts Early Detection, Cuts Missed Diagnoses
The trial tested the Transpara Detection AI system from ScreenPoint Medical, which was trained and validated on more than 200,000 breast exams from over 10 countries. Researchers found that in the two years following initial scans, women in the AI-supported group had a 12% lower rate of interval cancers—those diagnosed between regular screenings—compared to those screened with standard methods. Specifically, there were 1.55 interval cancers per 1,000 women in the AI group versus 1.76 per 1,000 in the control group.
Interval cancers are particularly worrisome because they often signal more aggressive or advanced disease and are linked to higher mortality. The MASAI results are striking: the AI not only reduced interval cancers, but also detected 27% fewer aggressive (non-luminal A subtype) cancers in the years following screening. Overall, the cancer detection rate for screen-detected cancers jumped by 29% with AI, and 81% of all cancers were caught at the screening stage—up from 74% in the standard-care group.
Radiologists’ Workload Drops, but Human Oversight Remains Key
Another advantage? The AI system lightened radiologists’ workload by 44% compared to the conventional double-reading approach, according to the study team led by Dr. Kristina Lång at Lund University. “AI-supported screening improves the early detection of clinically relevant breast cancers, which led to fewer aggressive or advanced cancers diagnosed in between screenings,” Dr. Lång said.
Despite these promising results, Dr. Lång and other experts stress that AI should act as a “second pair of eyes” for radiologists rather than replace them. While the findings suggest AI can help detect more cancers earlier and reduce pressure on overburdened healthcare providers, researchers urge further studies in different countries and health systems before adopting the technology widely.
Even so, this landmark trial suggests AI-assisted mammography could be a game-changer for breast cancer screening—helping catch more cancers early, reduce missed diagnoses, and give clinicians a much-needed boost in tackling the most common cancer in women worldwide.
