A study from the University of Cambridge, published by the Radiological Society of North America (RSNA), shows that an artificial intelligence tool called Mirai can identify women at higher risk of developing “interval” breast cancer—tumors that appear between regular screening mammograms. The research analyzed more than 134,000 mammograms from women aged 50 to 70 within the U.K.’s three-year screening program, detecting 524 cases of this type of cancer.
The deep-learning–based algorithm was able to retrospectively predict 42% of interval cancers among the 20% of women with the highest risk scores. According to the study, Mirai showed particularly strong performance in cancers diagnosed during the first year after a negative mammogram, although its effectiveness was lower in women with extremely dense breast tissue. Even so, it outperformed conventional prediction tools, paving the way for more personalized and effective screening programs.
The research team suggests that women with high scores could benefit from supplemental imaging techniques or shorter screening intervals. This approach could improve the early detection of aggressive tumors and reduce associated mortality.
According to the World Health Organization (WHO), breast cancer is the most common cancer among women worldwide and accounts for 12% of new cancer cases each year. The incorporation of artificial intelligence into screening programs is emerging as a promising avenue to optimize resources and tailor testing to individual risk, especially in high-demand public health systems.
Article translated from Diario Feminista
Diario Feminista
A newspaper whose pages feature news and contributions written from a feminist perspective, open to reading, commenting, and debate.