Google AI System Can Detect Breast Cancer Better Than Doctors

The system “outperformed” six radiologists in an independent study.

An AI system developed by Google can detect breast cancer better than doctors.

A study published on January 1, tested the accuracy of the AI system, which was developed by Google Health in collaboration with cancer researchers. It found that the system, which uses an algorithm based off of tens of thousands of mammograms from women in the U.S. and UK, could detect cancer more accurately than human radiologists.

The AI reduced false negatives by a little over 9% in the U.S. and nearly 3% in the UK. It also reduced false positives by almost 6% in the US and a little over 1% in the UK.

“In an independent study of six radiologists, the AI system outperformed all of the human readers,” the study explained. “This robust assessment of the AI system paves the way for clinical trials to improve the accuracy and efficiency of breast cancer screening.”

According to the American Cancer Society, mammograms miss roughly one in five breast cancers. About half of women getting annual mammograms over a 10-year period will also have a false positive at some point, which requires extra testing like MRIs or biopsies to conclude for sure whether cancer cells were detected.