Australian AI clears major milestone in bid to personalise breast cancer screening

An Australian-developed artificial intelligence (AI) program designed to improve the accuracy, personalisation and efficiency of breast cancer screening has reached a major milestone in the BRAIx clinical trials, as the technology expands across multiple clinical settings.

More than 45,000 BreastScreen Victoria clients have now participated in the randomised controlled trial, with over 21,000 mammograms assessed using the BRAIx AI reader. The study is now expanding to BreastScreen South Australia, marking another important milestone.

The BRAIx program is a partnership led by Associate Professor Helen Frazer from St Vincent’s BreastScreen and computational biologist Associate Professor Davis McCarthy from St Vincent’s Institute of Medical Research (SVI), together with collaborators from St Vincent’s Hospital Melbourne, The University of Melbourne, The University of Adelaide and BreastScreen Victoria. The program is funded by the Medical Research Future Fund (MRFF).

Associate Professor Frazer said the program aims to improve outcomes for women by enhancing current screening practices.

“Screening saves lives, but we can do it better,” she stated. “By combining expert radiologists with AI, we can detect cancers earlier, reduce unnecessary recalls and personalise screening based on risk.”

Results from simulation and evaluation studies have demonstrated the technology’s potential to improve both efficiency and accuracy. AI-assisted screening pathways reduced unnecessary recalls by up to approximately 10.9% and cut radiologist reading workload by around 50%, while maintaining or improving cancer detection. The BRAIx risk score also estimated four-year breast cancer risk more accurately than commonly used risk factors, with women in the highest 2% risk band having a 9.7% probability of diagnosis within four years despite a prior all-clear screening result.

The local approach of building sovereign AI capability by developing, training and validating the BRAIx algorithm based on Australian consumer data avoids dependence on international technology companies and proprietary algorithms, while strengthening Australia’s clinical and technical expertise in medical AI.

Associate Professor McCarthy said, “Training and validating BRAIx on Australian data means better accuracy, transparency and trust.”

In reimagining healthcare across the entire patient journey, Health Industry Hub^TM is the only one-stop-hub uniting the diversity of the Pharma, MedTech, Diagnostics & Biotech sectors to inspire meaningful change.

The Health Industry Hub^TM content is copyright protected. Access is available under individual user licenses. Please click here to subscribe and visit T&Cs here.