Breast cancer is the most common cancer affecting women worldwide. Early detection through regular mammograms saves lives, and programs like BreastScreen Victoria have helped reduce breast cancer deaths significantly. But there's always room to improve how we detect cancer, especially using new technologies like artificial intelligence (AI).
This project explored how AI could be used alongside human radiologists to improve breast cancer screening. The team tested different ways AI could be integrated into the current screening process, which usually involves two radiologists reviewing each mammogram, with a third radiologist review in if they disagree.
The research found:
- AI can help detect cancer more accurately when used as a second reader or when allowed to make decisions on high-confidence cases.
- These approaches improved cancer detection rates by up to 2.5% and reduce unnecessary follow-up appointments (false positives).
- Using AI could mean that fewer mammograms need to be read by humans, saving time and resources.
- The study also found that automation bias (where humans overly trust AI decisions) can sometimes reduce accuracy, especially in multi-reader settings.
S
St Vincent's Hospital
