Cardiovascular (CV) disease is the leading cause of death among women, however, is often under-recognised, underdiagnosed, and undertreated. In midlife when CV disease risk increases, 50% of Australian women aged 50-74 years annually attend mammography breast cancer screening. Repurposing routine mammography to also predict CV disease risk presents a unique opportunity to offer women a ‘two for one’ screening test for two of the leading causes of death (cancer and CV disease).
A novel sex-specific CV risk prediction algorithm based on routine mammographic images has been derived by our team and validated in an Australian cohort of almost 50,000 women. Using the recently validated sex-specific algorithm, this project aims to understand the acceptability and feasibility of implementing a mammography-based CV risk prediction tool in mammography centres in NSW. This will be achieved through two synergistic studies.
Firstly, I will explore the perspectives of primary care practitioners, mammography providers and women regarding using routine mammograms as a CV risk prediction tool in Australia. Then I will implement and assess the effectiveness, feasibility, acceptability and scalability of using our automated machine-learning mammographic image algorithm for CV risk prediction in mammography centres across NSW, Australia. The proposed work builds on previous research and established networks but will also involve new partnerships (eg. consumer groups, mammography centres and general practitioners). Throughout this 2-year fellowship, I will:
Last updated18 July 2025