Screening and AI: the BRAIx Project

Project Details

Researchers from the University of Melbourne’s Breast Cancer Unit are involved in the BRAIx project, a collaboration investigating the power of novel artificial intelligence (AI) techniques to transform breast cancer screening by improving analysis and interpretation of mammograms for Australian women.

Mammographic screening reduces the risk of dying from breast cancer, however, interpretation of mammographic images is challenging, subject to human variability, and has some room for improvement. Despite independent double reading of all mammograms by radiologists (and a third arbitration read if there is disagreement), approximately 33,000 Australian women are recalled annually for assessment and later determined not to have breast cancer (false positive), whilst approximately 1,000 women subsequently discover they have breast cancer after receiving an 'all clear' result (false negative). The cost of the public breast screening program, at over $300m annually, is also rising with Australia’s ageing population.

A unique cross-disciplinary health research team, BRAIx brings together clinicians, AI scientists and epidemiological and genomic researchers. The project aims to transform breast cancer screening using artificial intelligence (AI). The team has demonstrated the opportunity to significantly improve screening outcomes, lower harms and reduce costs using AI.

Researchers

Investigators

A/Prof Helen Frazer, Clinical Director, St Vincent’s Hospital Melbourne BreastScreen

The Late Professor John Hopper, University of Melbourne

A/Prof Davis McCarthy, St Vincent's Institute Medical Research

Professor Gustavo Carneiro, University of Surrey, UK

Dr Peter Brochie, St Vincent’s Hospital

Dr Jocelyn Lippey, St Vincent’s Hospital

Researchers

Dr Kevin Nguyen

Dr Osamah Al-Qershi

Mr Michael Elliott

Collaborators

St Vincent’s Hospital Melbourne, St Vincent’s Institute Medical Research, BreastScreen Victoria, University of Melbourne, Australian Institute of Machine Learning at the University of Adelaide.

Funding

Funded by the Medical Research Future Fund

Research Publications

Frazer HML, et al. ADMANI: Annotated Digital Mammograms and Associated Non-Image Datasets. Radiol Artif Intell. 2023;5(2):e220072.

Frazer HML, et al. Integrated AI reader development and evaluation provides clinically-relevant guidance for human-AI collaboration in population mammographic screening. medRxiv 2022.11.23.22282646.

Al-Qershi O, …., Hopper JL. AutoCumulus: an automated mammographic density measure created using artificial intelligence. medRxiv 2024.302158. (under revision at Breast Cancer Research)

Hopper JL, Nguyen TL, Elliott SM, … Frazer HML. BRAIx risk score: an automated mammogram-based biomarker for breast cancer created by applying artificial intelligence. (under revision at The Lancet Digital Health). https://ssrn.com/abstract=4764786