AI Sees Invisible Heart Danger Years Early

A pioneering artificial intelligence (AI) tool developed by University of Oxford researchers can now identify fat around the heart that is undetectable by the human eye. This AI examines routine heart CT scans for subtle textural changes, indicating inflammation and unhealthy heart muscle beneath. These changes are imperceptible through standard medical imaging tests.

The AI analysis serves as an early warning system, flagging patients at high risk of developing heart failure up to five years before symptoms manifest. The tool was trialed on 72,000 patients across nine NHS Trusts in England between 2007 and 2022. Findings indicate that individuals in the highest risk group were twenty times more likely to develop heart failure compared to those in the lowest risk group.

Published in the Journal of the American College of Cardiology, the study reveals the algorithm predicts future heart failure risk with 86% accuracy. Experts believe this tool has the potential to significantly impact cardiovascular care within the UK, where around 350,000 cardiac CT scans are performed annually. Researchers are also working to adapt the method for all chest CT scans, regardless of the reason for the scan, to broaden its application.

Dr. Sonya Babu-Narayan of the British Heart Foundation highlighted that heart failure is often diagnosed too late, by which point significant heart muscle damage may have occurred. This AI's ability to identify high-risk individuals allows for closer monitoring and earlier intervention, crucial for improving patient outcomes and longevity. Professor Charalambos Antoniades, leading the research, emphasized that the AI produces an absolute risk score without human input, representing a major step forward in treating heart failure.