CARA Lab a joint effort involving Abbott, RadboudUMC, and AmsterdamUMC, focuses on the analysis of optical coherence tomography (OCT) pullbacks to provide insight on the condition of coronary arteries. OCT pullbacks are high-resolution data streams, acquired without using ionizing radiation, that can show cross-sections of the vessels. Our goal is to develop trustworthy AI-systems for real-time analysis of OCT images to support clinical decision-making during coronary artery intervention to improve patient outcomes and to ensure the AI models are trustworthy: accurate, robust and reliable.

This lab is part of the 10-year LTP ROBUST program “Trustworthy AI-based Systems for Sustainable Growth” consortium, which unites 17 knowledge institutions, 19 participating industry sponsors and 15 civil-social organizations from across the Netherlands in conjunction with ICAI, the Innovation Center for Artificial Intelligence at the University of Amsterdam.