Researchers at the University of Oxford have created an innovative AI tool capable of predicting the likelihood of heart failure up to five years in advance.
Heart failure affects over 60 million individuals globally, a condition where the heart struggles to effectively circulate blood throughout the body. Experts believe that identifying this condition early could significantly enhance patient outcomes, enabling healthcare providers to manage the disease more effectively or potentially prevent its onset altogether.
The newly developed AI tool analyzes indicators in the fat surrounding the heart, which signify inflammation and poor heart health—details that are not discernible to the naked eye.
Prior to this advancement, there was no reliable method for forecasting heart failure through standard cardiac CT scans, according to the research team. The tool generates a risk score for patients, assisting healthcare professionals in making informed decisions about monitoring and treatment strategies.
Individuals identified as being in the highest risk category were found to be 20 times more likely to experience heart failure compared to those in the lowest risk category, with a roughly 25% chance of developing the condition within five years.
This AI 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. It demonstrated an impressive 86% accuracy rate in predicting the risk of heart failure over the next five years. The findings were published on Wednesday in the Journal of the American College of Cardiology.
Professor Charalambos Antoniades, a cardiovascular medicine expert at Oxford and the study’s lead, remarked, “We have leveraged advancements in bioscience and computational technology to significantly progress in heart failure treatment. Our AI tool can analyze cardiac CT scan data and generate a precise risk score for each patient autonomously. While this initial research focused on cardiac CT scans, we are now exploring the application of this approach to any chest CT scans conducted for various reasons.”
The Oxford research team is currently pursuing regulatory approval to implement this tool within healthcare systems, including the NHS, with hopes of integrating it into the routine analysis of cardiac CT scans in hospital radiology departments.
Dr. Sonya Babu-Narayan, clinical director at the British Heart Foundation, which provided funding for the study, stated, “Heart failure is often diagnosed too late, sometimes only upon hospital admission. A late diagnosis can result in significant damage to the heart muscle, which could potentially be mitigated. This tool has the potential to aid doctors in identifying heart failure earlier by allowing for closer monitoring of those at the highest risk. Timely diagnosis is essential, as it enables healthcare providers to manage a patient’s condition more effectively, thereby improving their chances of living longer and healthier lives. This study underscores the capability of technology to enhance cardiovascular care.”
Health experts emphasize that the best strategies for maintaining heart health include a diet rich in fruits and vegetables, regular physical activity, maintaining a healthy weight, quitting smoking, reducing alcohol intake, and managing blood pressure levels.
