AI-backed mammograms can detect heart disease in women: Study
What's the story
Artificial intelligence (AI) is being used to detect heart disease in women through routine mammograms, a new study has found. The research was presented at the European Society of Cardiology's annual congress, the world's largest heart conference. The technique could revolutionize healthcare for millions of women by detecting coronary heart disease, high blood pressure, and stroke during breast cancer screenings.
Detection rates
Study involved over 29,000 women
The study, conducted by Israeli doctors, analyzed 97,364 scans from 29,921 women.
They found that 16% of the participants had high blood pressure while 2.5% had coronary heart disease or a history of stroke.
A machine-learning model was trained to identify these conditions and proved successful in doing so with an accuracy of up to 86%.
Consistency
AI model showed consistent results across various demographics
The AI model proved reliable in identifying high blood pressure and coronary heart disease with accuracy rates of 79% and 78%, respectively.
Notably, these results remained consistent regardless of the women's age or their cancer status.
This consistency is a major breakthrough as it shows that the AI can be used on a wide range of women without any bias.
Future prospects
Cardiovascular disease often goes undiagnosed in women
Dr. Viana Copeland from Tel Aviv University, who presented the findings in Munich, said that cardiovascular disease (CVD) is often underdiagnosed and undertreated.
She noted that many women only seek medical help when their CVD is already advanced.
However, since routine breast cancer screenings are common among women, analyzing these scans for heart health could provide a scalable solution without requiring additional imaging tests.
Challenges ahead
Efforts underway to enhance AI model's capabilities
A team of doctors and researchers are working to improve the AI model's accuracy and expand its detection capabilities.
Elena Arbelo, an expert member of the European Society of Cardiology communication committee, called the findings "compelling."
She said, "The challenge now is to establish accuracy and reliability - to move from experimentation to clinical implementation."