Artificial intelligence could help detect the spread of breast cancer without a biopsy

A new study finds that artificial intelligence (AI) can help detect breast cancer that has spread to other parts of the body, without the need for biopsies.

AI analyzes MRI scans to detect the presence of cancer cells in the lymph nodes under the arms, researchers report.

In clinical practice, AI could help avoid 51% of unnecessary surgical lymph node biopsies to test for cancer, while correctly identifying 95% of patients who have breast cancer has spread, the study results showed.

Most deaths from breast cancer involve spread to other parts of the body, and the cancer usually first spreads to a lymph node in the armpit, explained lead researcher Dr. Basak Dogan, director of breast imaging research at the University of Texas Southwestern Medical Center..

Detecting cancer that has spread to a lymph node "is critical for making treatment decisions, but traditional imaging techniques are not sensitive enough to effectively detect it," Dogan said in a news release from the medical center.

Patients with benign MRI or needle biopsy results are often forced to undergo surgical lymph node biopsy because these tests may miss significant numbers of cancer cells that have spread beyond the breast, Dogan added.

Researchers trained the AI using MRI scans of 350 newly diagnosed breast cancer patients who had cancer found in the lymph nodes.

Testing has shown that the new AI is significantly better at identifying such patients than doctors using MRI or ultrasound, researchers reported in the journal Radiology: Imaging Cancer.

"This is an important breakthrough because surgical biopsies have side effects and risks, despite the low probability of a positive result confirming the presence of cancer cells," Dogan explained. "Improving our ability to exclude the presence of cancer cells in lymph nodes during routine MRI using this model may reduce this risk, improving clinical outcomes."