AI promises to eliminate unnecessary breast cancer biopsies

Editor's Note

An AI model that outperformed MRI and ultrasound in identifying patients with axillary breast cancer metastasis shows the technology’s potential to reduce the need for needle or surgical biopsies, according to developers at UT Southwestern Medical Center.

In a May 21 report on their new AI model, researchers note that “Most breast cancer deaths are due to metastatic disease, and the first site is usually an axillary lymph node.” However, traditional imaging techniques employed without AI “do not have enough sensitivity to rule out axillary metastasis.” Using machine learning techniques to train the model on images of dynamic contrast-enhanced breast MRI exams from 350 newly diagnosed breast cancer patients, this AI model showed the potential to prevent invasive procedures followed by surgery to remove and test whether axillary nodes harbor cancer cells.

“In clinical practice, the AI model would have helped avoid 51% of benign (noncancerous) or unnecessary surgical sentinel node biopsies while correctly detecting 95% of patients with axillary metastasis,” write the researchers, who published their findings in Radiology: Imaging Cancer. Using the model along with standard imaging exams also could eliminate the patient stress and expense accompanying additional testing.