Artificial intelligence, through the AIRE model, is transforming healthcare by anticipating disease and death risks. This innovation promises treatments before symptoms appear. Learn how science fiction dreams come true in health management!
Medical innovation – AI electrocardiograms that predict disease and premature death The truth
We live in an era where science fiction fantasies become reality and technological advances allow us to take a preventative approach to health management. Recently, researchers from Imperial College London and Imperial College Healthcare NHS Trust published a study in The Lancet Digital Health journal highlighting the potential of artificial intelligence-assisted electrocardiogram (ECG) (AIRE) – electrocardiograms with AI – in the early identification of heart disease and prioritization of urgent cases.
Trained on millions of ECGs, the AIRE model can predict with 78% accuracy the risks of death and the development of heart disease. Studies also suggest that the use of AI-enhanced (AI) ECGs can extend diagnostic capabilities beyond traditional methods, facilitating their integration into the UK public health system (NHS) for now.
The next step is to test the model under real clinical practice conditions, with studies planned for 2025. This medical innovation not only has the potential to improve patient outcomes, but also to optimize resource management within the healthcare system, offering a hope new in this field, even in the context of the ideal of eternal youth.
Innovation in cardiology: Artificial intelligence revolutionizes electrocardiography
Researchers from Imperial College London and Imperial College Healthcare NHS Trust believe the results of their study could be implemented across the NHS within the next five years. The work is entitled “Electrocardiogram assisted by artificial intelligence for the estimation of mortality and cardiovascular risks: An actionable, explainable and biologically plausible platform”.
An electrocardiogram (ECG) monitors the electrical activity of the heart and is among the most commonly used medical tests worldwide.
The research team used extensive data sets from international sources, including millions of ECGs previously performed in routine care, to train the artificial intelligence model to analyze the ECGs and accurately predict which patients developed new conditions, had an aggravation of existing diseases or subsequently died.
Electrocardiograms (ECGs) represent the flow of electrical signals that travel within and between the chambers of the heart, including the atria and ventricles. The AI model was trained to “read” this data and identify patterns in the electrical signals. According to the researchers, this model has the ability to perceive and interpret ECG patterns with a complexity and subtlety superior to that of a cardiologist.
AIRE: 78% accuracy in 10-year risk prediction
Dr Arunashis Sau, clinical academic lecturer at the National Heart and Lung Institute at Imperial College London and resident in cardiology at Imperial College Healthcare NHS Trust, who led this research explained: “We cardiologists rely on our experience and standard guidelines when analyzing ECGs, classifying them into «normal» and «abnormal» patterns to facilitate disease diagnosis. However, the AI model detects much finer details, allowing it to ‘spot’ problems in ECGs that would appear normal to us, and possibly long before the disease is fully manifested.”
The artificial intelligence model, known as ECG AI Risk Estimation (AIRE), was able to correctly determine the risk of death over 10 years in 78% of cases, classifying the risk from high to low. As for the cases where the model got it wrong, the researchers suggest that these errors could be influenced by unpredictable factors, such as the patient’s subsequent treatments or unexpected causes of death.
The system can predict future health risks, including heart rhythm problems, heart attacks and heart failure, and when a person might die even from non-heart causes. The researchers found that the system predicts these risks with a high degree of accuracy.
Artificial intelligence transforms ECGs: detecting signs of aging
Dr. Arunashis Sau explained: “ECGs collect a significant amount of information from the whole body, because diseases such as diabetes, which affect organs such as the kidneys or liver, also have an impact on the heart. Our analysis shows that artificial intelligence can provide information not only about the heart, but also about the condition of other parts of the body, thus having the potential to detect accelerated aging.”.
AI-enhanced ECGs are already known for their high accuracy in diagnosing heart disease, but have not yet been used to inform clinicians about patients’ specific risk of developing a range of treatable diseases in the future. Currently, these ECGs are not part of routine care or diagnostic practice in hospitals.
The researchers also analyzed imaging as well as genetic information, which allowed them to confirm that the predictions made by the artificial intelligence correlated with real biological factors in the structure and function of the heart. They point out that this is critical to the credibility of the model, demonstrating its ability to detect subtle changes in the structure of the heart over time that can represent early signs of risk of disease or death.
Clinical trials in 2025: AI promises to transform patient care
The study’s lead author, Dr Fu Siong Ng, lecturer in cardiac electrophysiology at the National Heart and Lung Institute at Imperial College London, said: “Our research has shown that this AI model is a credible and reliable tool that could be programmed in the future to be used in various areas of the NHS. , providing physicians with relevant risk information. This could have a positive impact on how patients are treated, helping to improve their longevity and quality of life. In addition, it could reduce waiting lists and facilitate a more efficient allocation of resources. We believe this model could bring significant benefits both for the NHS and globally”.
The doctor, who is also a consultant cardiologist at Imperial College Healthcare NHS Trust and Chelsea and Westminster Hospital NHS Foundation Trust, added: “The next critical step is to test whether implementing these models can actually improve patient outcomes in clinical trials.”
Clinical trials for AIRE are already planned at hospitals within Imperial College Healthcare NHS Trust and Chelsea and Westminster Hospital NHS Foundation Trust. These studies will focus on evaluating the benefits of implementing the model among real patients and will start by mid-2025. Patients will be recruited from outpatient clinics as well as inpatient wards.
The future of health: continuous monitoring through wearable devices
Professor Bryan Williams, Scientific and Medical Director of the British Heart Foundation, said: “This large and exciting study provides insight into how artificial intelligence could improve the diagnosis of heart disease. ECGs have been used to assess the heart for more than a century, and this research demonstrates the remarkable power of artificial intelligence to extract essential health information from a routine test.”
Director Williams noted that the medical world is looking forward to the implementation of artificial intelligence in routine clinical practice and its impact on accelerating and informing decision-making so that patients benefit from the fastest and most effective treatments.
Dr Sau added: “Now we need to evaluate the performance of the model in a real health system. In the future, it may be possible to equip patients with wearable technology that allows doctors to perform continuous remote monitoring and activate an alert system.”
