AI-APPROVE: Smarter Use of Heart Scans, Stronger Societies

AI-APPROVE: A Clinical Decision Support System for Optimal Deployment of Coronary Imaging

Within the AI-APPROVE project, we aim to improve the exclusion diagnosis of coronary artery disease (CAD), a heart condition that is becoming increasingly common in the Netherlands. It is a collaboration between the University Medical Center Utrecht, MedxAI, Expertise Center Care Algorithms and the St. Jansdal Hospital. This collaboration is all about bringing together parties from the public and private sectors to make great strides in cardiac healthcare.

Currently, cardiologists often use cardiac scans such as CCTA to rule out CAD in patients with CAD-related symptoms, but many of these scans may not be necessary. With CAD expected to affect nearly one million people in the Netherlands by 2040, it is important to find ways to rule out CAD more efficiently, preferably without cardiac imaging. AI-APPROVE is working on this by trying to reduce the number of unnecessary cardiac scans. This means fewer procedures for patients, less exposure to radiation and contrast agents, and more efficient use of available resources as the pressure on care is high and even getting higher.

At the heart of AI-APPROVE is a tool called the ARGUS algorithm. It uses AI to predict whether an individual really does not need a heart scan. Previous studies have shown that ARGUS is very good at identifying individuals who do not have CAD, allowing them to avoid unnecessary scans. The project is now further validating ARGUS and integrating it into the healthcare system. We aim to see how it influences cardiologists’ decisions about heart scans and get it ready for use in the clinic.

AI-APPROVE fits exactly with the Dutch Ministry of Health's goals of helping people live healthier lives and reducing inequalities in health care. By bringing ARGUS to hospitals and clinics, AI-APPROVE not only changes how we rule out heart disease; it also shows how AI can play a major role in our healthcare system and set the stage for more advances in the future.

Summary
AI-APPROVE focuses on improving the diagnostic process of coronary artery disease (CAD), particularly in low- to intermediate-risk individuals. The cornerstone of the project is the development of the ARGUS algorithm, an AI tool developed to assess the need for coronary imaging, such as coronary computed tomography (CCTA), in patients with CAD-related symptoms. The initiative, a collaboration between the University Medical Centre Utrecht, MedxAI and the Dutch Heart Foundation, aims to reduce the frequency of unnecessary coronary imaging, thereby reducing patient risk and increasing healthcare efficiency. The project includes the integration of ARGUS into a data platform developed by Expertisecentrum Zorgalgoritmen and three clinical studies to evaluate the impact on diagnostic decisions by cardiologists. AI-APPROVE aligns with the Dutch Ministry of Health's mission by contributing to healthier life expectancy, reducing health disparities and optimising healthcare resources. The project promises an important step towards improved diagnostic accuracy for CAD and a reduction in unnecessary medical procedures, leading to more efficient and equitable healthcare outcomes.
Technology Readiness Level (TRL)
3 - 5
Time period
36 months
Partners