Percutaneous coronary intervention (PCI) is a minimally-invasive procedure for treating patients with coronary artery disease. PCI is typically performed with image guidance using X-ray angiograms (XA) in which coronary arteries are visualized with X-ray opaque contrast agents. During the PCI procedures, image guidance and visualization are often bottlenecks in the feedback loop of the physician, medical instruments and the patient. In this thesis we have reported on the development and evaluation of dynamic image analysis approaches for improving image guidance for PCI. We have proposed layer separation methods to enhance the visibility of coronary arteries in XA. Two applications have also been demonstrated to benefit from layer separation for detection of respiratory signals and contrast inflow in XA series. Additionally, we have reported the development of a dynamic coronary roadmapping approach for providing real-time visualization of the coronary arteries, even after the contrast agents have disappeared from the arteries. Both approaches have the potential to reduce the use of contrast agent during PCI procedures. The techniques developed in thesis could potentially be employed to build an improved image guidance system that can help interventional cardiologists to view relevant information that is needed for PCI.

Medical imaging, Dynamic analysis, X-ray angiography, Coronary artery, Percutaneous coronary interventions, Deep learning, Machine learning, Computer vision
W.J. Niessen (Wiro) , T.W. van Walsum (Theo)
Erasmus University Rotterdam
Financial support by the Dutch Heart Foundation for the publication of this thesis is gratefully acknowledged
ASCI dissertation series
Department of Radiology

Ma, H. (2020, February 12). Dynamic Analysis of X-ray Angiography for Image-Guided Coronary Interventions (No. 410). ASCI dissertation series. Erasmus University Rotterdam. Retrieved from