Oxford AI detects early heart failure risk in routine CT scans with 86% accuracy in 72,000 patients

by
Alyssa Davidson

Posted: Apr 10, 2026 10:37 AM Updated: Apr 10, 2026 10:38 AM

by Anastasia Oh

Correction and fact check date: April 10, 2026, 10:37 AM

We sometimes use automatic translation plugins to improve your local language experience. The automatic translation may not be accurate, so please read the original text for accurate information.

Oxford University researchers have developed an artificial intelligence system that can predict a patient’s risk of developing heart failure up to five years in advance, achieving a verification accuracy of 86% on more than 72,000 patients. This approach does not require additional testing, specialist intervention, or new medical equipment because it relies on cardiac CT scans already routinely performed in clinical practice.

The study, led by Professor Charalambos Antoniades and published in the Journal of the American College of Cardiology, addresses a long-standing limitation in cardiology. Heart failure is typically diagnosed only after significant structural damage has already occurred, at which point prevention options are often limited. The proposed system shifts attention to early biological changes that appear years before noticeable symptoms appear.

At the heart of the model is an unconventional data source: fat around the heart, known as pericardial adipose tissue. Although traditionally overlooked in routine scan analysis, this tissue appears to reflect underlying inflammatory and metabolic changes occurring in the heart muscle itself.

According to the researchers, these fatty deposits gradually change their texture in response to stress on the cardiovascular system, creating patterns that are undetectable through standard human interpretation of imaging results. The AI ​​system is designed to identify these subtle changes and translate them into quantifiable risk estimates for future heart failure.

Reading signals invisible to the human eye

Cardiac CT imaging is widely used by the UK National Health Service to investigate chest pain and assess coronary artery disease, with hundreds of thousands of scans performed each year. In a typical clinical workflow, radiologists primarily focus on arterial occlusions and visible abnormalities, while analytical attention to surrounding fatty tissue is limited.

The Oxford model repurposes overlooked layers of data by analyzing histological features within pericardial fat. Using machine learning techniques trained on anonymized CT data from over 59,000 NHS patients, the system learned to associate specific imaging patterns with the development of future heart failure over a long-term follow-up period.

In validation testing with 13,424 additional patients, the model was 86% accurate in predicting 5-year heart failure risk. People classified in the highest risk group were found to be around 20 times more likely to develop the condition than those in the lowest category, with an estimated one in four chance of developing it within five years.

Importantly, the system automatically generates a risk score without the need for manual input from clinicians. It does not replace existing diagnostic processes, but is positioned as a potential decision support tool.

From heart scans to chest CTs – and the road to the NHS

The broader goal of the research is to expand the technology beyond cardiac-related imaging. The team is currently working on adapting the model to analyze standard chest CT scans, including scans used for lung cancer screening and respiratory diagnostics. Given the much larger volume of chest CT images compared to heart-related scans, this adaptation could significantly increase the reach of the system.

Clinically, its implications are relevant for early intervention. By identifying high-risk patients years before symptoms appear, health care providers can tailor monitoring strategies, initiate preventive care earlier, and prioritize resources more effectively. With more than a million people already affected by heart failure in the UK, the potential impact on long-term healthcare needs is significant.

Plans are currently underway to seek regulatory approval for integration into routine radiology workflow within the NHS. Once adopted, the system will work in the background of standard imaging procedures to generate automated risk assessments without additional costs or changes to scanning protocols.

The study was supported by the British Heart Foundation and the National Institute for Health and Care Research Biomedical Research Center in Oxford. This reflects a broader shift in medical imaging, where artificial intelligence is increasingly being used not only to detect existing diseases but also to infer future risk from subtle and previously underutilized biological signals contained in routine scans.

As a dedicated journalist at MPost, Alisa specializes in the broad areas of cryptocurrency, AI, investing, and Web3. With a keen eye for new trends and technologies, she provides comprehensive coverage to inform and engage readers about the ever-evolving digital financial landscape.

• 
  
  

more articles