Australian researchers, in collaboration with the University of New South Wales and the University of Sydney, were the Independent Global Medical Institute and the George Global Health Institute, developed a machine learning model that allows mammographic images to be analyzed to successfully predict the risk of cardiovascular events in women.
Cardiovascular diseases, including coronary artery disease, arrhythmia, and heart valve disease, as well as cardiovascular events such as heart attacks and strokes, are the leading causes of death worldwide, and are almost approximately. 20 million deaths Every year.
Around 9 million women die from cardiovascular disease each year, and despite these high numbers, several international studies have shown that the symptoms of cardiovascular disease and cardiovascular events, and their risk factors, are far more overlooked in women than in men.
For example, a 2024 Research Women hospitalized for a heart attack were less likely to receive the necessary treatment and were more likely to die than men.
For this reason, researchers at the George Institute for Global Health wanted to find ways to use existing data to predict the risk of cardiovascular events in women.
“It is a common misconception that cardiovascular disease primarily affects men, leading to underdiagnosis and low treatment in women,” explained co-author Claire Arnott, PhD, PhD, associate professor and global director of the George Institute.
“By integrating cardiovascular risk screening through breast screening and mammogram use, many women are already involved in the stages of increased cardiovascular risk, allowing them to simultaneously identify and potentially prevent two major causes of illness and death.”
Their research has been published in the journal heart title “Predicting cardiovascular events from everyday mammograms using machine learning” They developed a fully automated, deep learning algorithm that can analyze the architecture and characteristics of the whole breast to predict cardiovascular risks in women undergoing routine mammography screening for breast cancer.
Ideas to use Mammogram image Understanding this risk is not new, but previous research has focused only on some features of mammographic images, such as mammographic artery calcification. However, this is limited because the risk of cardiovascular events can arise from many factors. For example, calcification of the breast artery cannot be applied accurately to older women.
In their study, the researchers looked at mammographic data from 49,196 women aged 35-94, with an average age of about 60 years and a median follow-up of 8.8 years. Of these women, 3,392 reported experiencing the first major cardiovascular event during follow-up, including atherosclerosis, heart failure, heart attacks and stroke.
Researchers trained an automated algorithm to analyze the full range of internal breast structures and characteristics of these routine mammogram images, taking into account the age of women and predicting the risk of major cardiovascular disease over a decade.
The AI model was then compared with other risk scores and calculators. These calculators require multiple data points based on known cardiovascular risk factors, including blood pressure and cholesterol.
“We found our model to work in a similar way without the need for extensive clinical and medical data,” Arnott said. “Our model is the first to use the various features of mammographic images simply combined with age. A key advantage of this approach is that it does not require additional history or medical record data, and requires less resource implementation, but is still very accurate.”
As a next step, researchers aim to validate algorithms in a more diverse patient population, using a variety of screening practices to assess and further refine the generalizability of AI models.
