AI Boosts Breast Cancer Detection in World-First Trial

A world-first trial in Sweden shows that artificial intelligence (AI) can help doctors detect more cases of breast cancer during routine scans.

The study, published in The Lancet, is the first completed randomised controlled trial testing AI-assisted breast cancer screening. Researchers say the findings support wider use of AI to help overworked radiologists.

The trial included over 100,000 women who had routine mammograms in 2021 and 2022. They were randomly divided into two groups: one where a single radiologist was assisted by AI, and another following the standard European method of two radiologists reading the scans.

Results showed the AI-supported group detected 9% more cancers than the traditional method. Over the next two years, this group also had a 12% lower rate of “interval cancers,” which are cancers diagnosed between routine screenings and can be especially dangerous. The improvement was consistent across age groups and breast density, and the rate of false positives remained similar.

Kristina Lang, lead author from Lund University, said, “Widespread use of AI in breast cancer screening could reduce radiologists’ workload and help detect more cancers early, but implementation should be cautious and closely monitored.”

Experts stress that AI cannot replace doctors. Jean-Philippe Masson from the French National Federation of Radiologists said, “The radiologist’s experience must correct AI results. AI may flag changes that aren’t actually cancer.”

Stephen Duffy, professor at Queen Mary University of London, added the study confirms AI-assisted screening is safe, though he noted the reduction in interval cancers is not yet significant and further follow-up is needed.

The AI system, called Transpara, was trained on more than 200,000 scans from 10 countries. Earlier results showed it can nearly halve the time radiologists spend reading scans.

According to the World Health Organization, over 2.3 million women were diagnosed with breast cancer and 670,000 died from the disease in 2022, highlighting the need for early detection.