The study randomly assigned women either to standard mammogram reading or to mammogram assessment supported by artificial intelligence. The results showed that the AI-supported group had a significantly lower proportion of stage III and stage IV breast cancers compared with the control group, whose images were interpreted exclusively by radiologists without AI assistance.
The authors emphasize that the key difference was a higher sensitivity for detecting early-stage disease in women whose images were assessed with AI support. The AI system helped identify subtle abnormalities in mammographic images that could have been missed during conventional reading, leading to earlier cancer diagnosis. As a result, advanced-stage disease—typically associated with poorer prognosis and a higher treatment burden—was observed less frequently in the AI group.
The study highlights that earlier detection of breast cancer is critical for treatment effectiveness, as cancers diagnosed at stage I or II can often be treated less aggressively and are associated with higher survival rates than those detected at later stages. The use of artificial intelligence in mammography may therefore contribute to improved clinical outcomes and to a reduction in the number of cases requiring intensive therapy.
These findings are particularly important because they are based on a randomized controlled design, which is considered the gold standard for evaluating the effectiveness of medical interventions. Previous research on AI in medical imaging has often relied on retrospective or comparative analyses, which did not provide equally robust evidence of AI’s effectiveness in real clinical settings.
The authors note, however, that although the results are encouraging, further studies involving larger patient populations and different healthcare systems are needed to confirm whether similar benefits of AI-supported mammography can be observed more broadly in routine clinical practice. Wider implementation will also depend on factors such as technology costs, access to high-quality data, and regulatory frameworks governing the use of artificial intelligence in diagnostic care.