Researchers at Oxford University have developed an innovative AI tool that can forecast the likelihood of heart failure up to five years before it occurs.
Currently, over 60 million individuals globally experience heart failure, a condition where the heart struggles to effectively pump blood throughout the body. Identifying cases before they escalate into full-blown heart failure represents a significant advancement, as it allows medical professionals to better prepare for and manage the condition, potentially preventing it altogether.
This AI tool analyzes indicators found in the fat surrounding the heart that signal inflammation and poor health—details not discernible to the naked eye.
Previously, there was no reliable method to anticipate heart failure using standard cardiac CT scans. The AI tool provides a risk assessment for patients, helping physicians determine the appropriate level of monitoring and care.
Individuals categorized in the highest risk bracket were found to be 20 times more likely to develop heart failure compared to those in the lowest risk category, with roughly a 25% chance of experiencing the condition within five years.
This groundbreaking tool was developed using data from 72,000 patients across nine NHS trusts in England, who were monitored for a decade following their CT scans. The AI demonstrated an 86% accuracy rate in predicting heart failure risk over the next five years. The findings were published in the Journal of the American College of Cardiology on Wednesday.
Charalambos Antoniades, a professor of cardiovascular medicine at Oxford and the study’s lead, stated: “We have leveraged advancements in bioscience and computing to make significant strides in heart failure treatment.
“Our new AI tool can interpret cardiac CT scan data and generate a precise risk assessment for each patient without any human intervention. While this study focused on cardiac CT scans, we aim to adapt this approach for any chest CT scan conducted for any purpose.
“Such advancements will empower healthcare providers to make better-informed choices regarding patient treatment, ensuring that those at highest risk receive the most comprehensive care.”
The Oxford team is currently seeking regulatory approval to implement this tool across healthcare systems, including the NHS. They aspire to integrate it into routine analysis of cardiac CT scans in hospital radiology departments.
Dr. Sonya Babu-Narayan, the clinical director at the British Heart Foundation, which funded the study, remarked: “Heart failure is often diagnosed too late—sometimes only upon hospital admission. Late diagnosis can result in significant and avoidable damage to the heart muscle. This tool could facilitate earlier detection of heart failure by enabling closer monitoring of those at greatest risk. Early diagnosis is essential as it allows for better management of the condition, ultimately improving patients’ chances of living longer and healthier lives. This study highlights the immense potential of technology to enhance cardiovascular care.”
Experts emphasize that maintaining heart health is best achieved through lifestyle choices such as consuming ample fruits and vegetables, staying physically active, maintaining a healthy weight, quitting smoking, reducing alcohol intake, and managing blood pressure effectively.
