Scientists from Australian universities and The George Institute for Global Health have developed an artificial intelligence tool that uses mammogram images and a woman’s age to predict her risk of hospitalization or death from heart issues within the next 10 years.
The tool was created to address the significant lack of routine heart disease screening for women, despite cardiovascular diseases being the cause of 35% of female deaths globally.
The study was published in the journal Heart on September 17, 2025.
Leveraging existing screenings to close a healthcare gap
The AI model analyzes mammograms, which are already performed regularly for breast cancer detection, to assess heart health. This approach avoids the need for separate cardiac examinations. The development was driven by the fact that many women are unaware of their cardiovascular risks due to a common, persistent misconception.
“It’s a common misconception that [heart disease] predominantly affects men, resulting in underdiagnosis and undertreatment of the condition in women.”
By integrating heart risk assessment into routine breast cancer screening, the tool provides a two-for-one evaluation in a single appointment, enabling earlier and more efficient preventative care.
An accurate and resource-efficient model
The model was trained and validated on data from over 49,000 women. Researchers found that its predictions were as accurate as existing models that require additional clinical information, such as blood pressure and cholesterol levels. This makes the new tool highly efficient and easier to implement.
“The key advantage of the model is that it doesn’t require additional history taking or medical record data, making it less resource intensive to implement, but still highly accurate.”
Potential for remote and underserved communities
The tool has significant potential for use in regions with limited access to medical facilities. Dr. Jennifer Barraclough, a study author, highlighted its applicability for services like mobile mammography units in rural Australia. These units could provide both breast cancer and heart health screenings to remote populations without needing additional infrastructure.
Next steps for the research
The research team now plans to test the model in more diverse populations to ensure its reliability and identify any potential barriers to widespread implementation, such as technical or regulatory challenges.
“We have shown the potential of this innovative new screening tool, so we now look forward to testing the model in additional, diverse populations and understanding potential barriers to its implementation.”
