Level Set Based Integration of Segmentation and Computational Fluid Dynamics for Flow Correction in Phase Contrast Angiography

摘要

A novel approach to correct flow data from phase contrast angiography (PCA) is presented. The method is based on combining computational fluid dynamics (CFD) and segmentation in a level set framework. The PCA-MRI velocity data is used in a partial differential equation (PDE) based level set method for vessel segmentation, and a second level set equation solving for a physically meaningful flow. The second level set is implemented using the ghost fluid method, where the MR data defines initial and boundary conditions. The segmentation and CFD systems are simultaneously integrated to provide a robust method yielding a physically correct velocity and optimal vessel geometry. The application of this system to both synthetic and clinical data is demonstrated and its validity is discussed.