Oxford's AI Predicts Heart Failure 5 Years in Advance

A Major Breakthrough in Early Detection

Researchers at the University of Oxford have developed an artificial intelligence (AI) system that could transform the way heart failure is detected. This technology promises to predict the disease up to five years before the first symptoms appear, providing a valuable intervention window for at-risk patients.

Heart failure is a feared condition that affects tens of millions of people worldwide. It often occurs without warning signs, making its management complex. However, thanks to this new approach developed in England, it is now possible to anticipate the risk well in advance, which could be a game changer for many patients.

How the AI Works

The project, originating from the University of Oxford, relies on a sophisticated algorithm that analyzes traditional cardiac scans. Unlike conventional methods, this AI focuses on the fatty tissue surrounding the heart. This often-overlooked tissue holds crucial clues such as inflammation and subtle anomalies that the human eye cannot perceive.

By evaluating these signals, the AI assigns a risk score to each patient. This score allows doctors to better target their monitoring and adjust treatments accordingly. The algorithm has been trained on a vast database comprising 72,000 patients followed over a ten-year period, with information collected from various NHS facilities.

The results are impressive: the AI achieves an 86% accuracy rate in predicting five-year risks. For patients identified as high-risk, the likelihood of developing heart failure is up to 20 times higher. In this group, one-quarter of patients could be affected within the next five years, which serves as an important alert signal for medical teams. These results have been published in the Journal of the American College of Cardiology, a leading reference in the field.

Often Late Diagnosis

Currently, the diagnosis of heart failure often comes too late. In many cases, the disease is only detected at the time of hospitalization, when the heart muscle is already damaged. According to the British Heart Foundation, this delay significantly limits the available therapeutic options.

The introduction of this AI could reverse this trend. By identifying at-risk patients well before symptoms appear, doctors can intensify monitoring and proactively adjust treatments. While this tool does not replace the expertise of physicians, it sharpens their judgment and allows them to focus resources on the most vulnerable patients.

Towards Integration into Healthcare Systems

One of the major advantages of this technology is its automation. The AI can analyze images without human intervention, thus easily integrating into existing workflows in radiology departments. Researchers are already working to adapt this technology so that it can be applied to all chest scans, even those performed for other medical reasons.

The Oxford team is now aiming for approval from health authorities to integrate this tool into healthcare systems. The goal is to add this analysis to routine examinations, thereby enriching the available data without altering current practices. If approval is obtained, deployment could begin within the NHS, with the potential for rapid adoption by other healthcare systems around the world.