www.medical-devices.tech
02
'26
Written on Modified on
AI-Enabled Computed Tomography Workflow for Coronary Artery Disease
Royal Philips and consortium partners deploy non-invasive clinical decision support infrastructure to optimize cardiac diagnostics and reduce unnecessary catheterization laboratory procedures.
www.philips.com
The COMBINE-CT European consortium, coordinated by Royal Philips, has commenced clinical validation of an artificial intelligence (AI)-supported coronary computed tomography angiography (CCTA) care pathway for coronary artery disease (CAD). The collaborative initiative integrates automated image processing, decision-support software, and advanced computed tomography (CT) architectures into cardiovascular healthcare workflows.
Operational Challenges in Cardiovascular Diagnostics
Coronary artery disease remains a primary cause of mortality, driven by atherosclerosis obstructing blood supply to myocardial tissue. Conventional diagnostic pathways rely on invasive coronary angiography within a cardiac catheterization laboratory. This invasive procedure requires catheter insertion and contrast media injection under X-ray imaging. Clinical records indicate that approximately 60% of patients undergoing diagnostic invasive coronary angiography show no hemodynamically significant stenosis. This creates operational bottlenecks in catheterization facilities, drives up institutional expenditure, and exposes non-obstructive patients to procedural intervention risks.
Consortium Architecture and Division of Roles
Addressing these procedural constraints required a public-private partnership combining industrial medical device engineering, pharmaceutical domain data, clinical infrastructure, and regulatory guidance. The initiative is co-funded with a 10 million euro budget through partner resources and the European Union Innovative Health Initiative (IHI).
Industrial partners Royal Philips and Novo Nordisk provide imaging modalities, informatics architecture, and metabolic risk modeling. Clinical and academic institutions — including Amsterdam UMC, Cardiologie Centra Nederland, Université Claude Bernard Lyon 1, Hospices Civils de Lyon, Sheba Medical Center, IBSAL, CIBER, and UNIKLINIK KÖLN — manage data acquisition and trial protocols across six European countries. Non-profit patient group EUPATI provides input on pathway accessibility.
Technical Integration and Multicenter Validation
The integrated system transitions standard CCTA from an observational tool into an automated diagnostic and interventional planning platform. The pipeline applies machine learning algorithms to assess arterial plaque composition, quantify stenosis severity, and simulate non-invasive coronary physiology. The workflow directly interfaces with radiology departments and cardiac catheterization laboratories to guide percutaneous coronary interventions (PCI).
Validation proceeds across five dedicated multicenter clinical trials:
- CONVENE: Prospective evaluation of the connected end-to-end diagnostic and procedural guidance pathway.
- CODEX-1: Retrospective assessment analyzing diagnostic algorithm accuracy across a dataset exceeding 1,000 patients.
- IMPACT: Comparative analysis correlating AI-enhanced CCTA results against invasive physiologic and intracoronary imaging metrics.
- SPCCT-STENT: Clinical evaluation using Spectral Photon-Counting CT, leveraging energy-resolving detector hardware to inspect post-PCI stented vessels and surrounding myocardium.
- EVOLVE: Serial CCTA study tracking plaque progression parameters in high-risk patients with type 2 diabetes mellitus.
Replacing elective diagnostic catheterizations with automated non-invasive stratification aims to preserve catheterization suites for therapeutic interventions, optimize clinical throughput, and standardize diagnostic thresholds across regional cardiac centers.
Edited by Evgeny Churilov, Induportals Media - Adapted by AI.
www.philips.com
Edited by Evgeny Churilov, Induportals Media - Adapted by AI.
www.philips.com
