A new artificial intelligence model designed to assess patients' risk of disease and early death will undergo trials in NHS hospitals starting in 2025. The AI program, known as AI-ECG risk estimation, or Aire, utilizes electrocardiogram (ECG) readings to identify potential health issues that may not be visible to doctors.
The trials will take place at Imperial College Healthcare NHS Trust and Chelsea and Westminster Hospital NHS Foundation Trust, with plans for additional sites to be confirmed. The initiative aims to process every ECG conducted in hospitals through the AI model, providing clinicians with critical insights into patients' health risks.
Dr. Fu Siong Ng, a consultant cardiologist at Imperial College, emphasized that the trials will initially focus on validating the accuracy of the AI in diagnosing conditions for patients already receiving ECGs. The ultimate goal is to enable early interventions based on AI predictions, potentially improving patient outcomes.
According to preliminary research published in Lancet Digital Health, the Aire model demonstrated an ability to accurately predict patients' risk of death over a decade in 78% of cases. It also showed a 79% accuracy rate in forecasting future heart failure and a 76% rate for serious heart rhythm problems.
Dr. Arunashis Sau, a clinical research fellow at Imperial College, highlighted the AI's potential to identify risks that human clinicians may overlook. The project aims to enhance patient management through better risk assessment, ultimately leading to improved preventative care.