Written By |
The program, known as AI-ECG Risk Estimation (AIRE), has demonstrated an ability to predict a patient’s risk of death within 10 years with up to 78% accuracy, according to studies
Search
HEALTH - Hundreds of British hospital patients may soon have their estimated time of death predicted by an AI-powered system. Using data from a single electrocardiogram (ECG) test — a simple procedure that records the heart’s electrical activity — the AI is designed to detect hidden health issues that doctors might miss.
The program, known as AI-ECG Risk Estimation (AIRE), has demonstrated an ability to predict a patient’s risk of death within 10 years with up to 78% accuracy, according to studies. AIRE will be trialled at two NHS trusts in London starting mid-next year, with hopes that it will be rolled out across the UK’s healthcare system within five years.
Dr. Arunashis Sau, a cardiology registrar at Imperial College Healthcare NHS Trust — one of the participating hospitals — emphasized that AIRE is not meant to replace doctors, but to perform tasks beyond human capabilities. “The goal here was to do something superhuman,” Dr. Sau explained.
The AI analyzes ECG results, detecting patterns in the heart’s electrical signals and genetic data to identify potential issues, such as heart rhythm problems or heart failure, before they become fully developed. It then provides a risk prediction, measured in years, for each patient.
Dr. Sau added, “The aim is to use the ECG to identify people at higher risk, who may benefit from further tests that could reveal more about their condition. While ECGs are common and inexpensive, they could be used to guide more detailed testing, potentially changing how we manage patients and reduce their risk.”
He continued, “AI can pick up details about heart structure and function, even detecting genetic information, which is something doctors can’t do by simply looking at heart tracings.”
'The goal is for every patient undergoing an ECG in a hospital to be assessed using the AI model'
The initial trial, also involving Chelsea and Westminster Hospital NHS Foundation Trust, will recruit several hundred patients, with plans to scale up for future studies. Research published in The Lancet Digital Health found that AIRE predicted future heart failure in nearly 80% of cases. In the study, the team trained the AI using 1.16 million ECG results from 189,539 patients.
The AI detected future serious heart rhythm problems in 76% of cases and predicted atherosclerotic cardiovascular disease — where arteries narrow, restricting blood flow — with 70% accuracy.
Dr. Fu Siong Ng, a consultant cardiologist at Imperial College Healthcare NHS Trust, outlined the future vision: “The goal is for every patient undergoing an ECG in a hospital to be assessed using the AI model. Clinicians will not only receive a diagnosis but also predictions about a range of potential health risks, allowing us to intervene early and prevent disease.”
For example, if the AI predicts a high risk of heart rhythm problems, doctors could take more aggressive preventative steps, such as weight loss programs or early medical treatments to stop conditions from worsening. Clinical studies are planned to explore the full potential of these interventions.
The trial comes amid alarming data showing premature deaths from cardiovascular problems, such as heart attacks and strokes, have reached their highest level in over a decade. Rising obesity rates, along with conditions like high blood pressure and diabetes, are believed to be major contributors. Delays in ambulance response times and long waits for treatment are also cited as factors.
Despite concerns from some, cardiologists have dismissed claims that Covid-19 vaccines are linked to an increase in heart problems.
Source: Daily Mail
