Wahle A, Ramaswamy SD, Olszewski ME, Rossen JD, Lopez JJ, Lai YG, Chandran KB, Sonka M:


Temporal Analysis of 3-D Coronary Plaque Morphology and Hemodynamic Shear Stress Distribution In-Vivo.

In:

Niederlag W, Lemke HU (eds):

Advances in Medical Imaging (I).

Health Academy

Number 2, Page 25-31, 2002

Republished from a presentation at the CARS-2002 conference (Images) (Links)

Abstract: The study of the progression of cardiovascular diseases as a function of hemodynamics and plaque morphology in human vasculature has gained recent interest. Previous studies of computational hemodynamics have implicated regions of flow reversal and separation, as well as relatively low wall shear stresses and oscillating shear stresses, as causes for initiation and growth of atherosclerotic plaques. Time-dependent vessel movement, changes of vessel curvature, torsion, rotation, lumen diameter, and length influence coronary hemodynamics. Consequently, determination of endothelial shear stresses in the coronary vessels is highly complex. To perform computational hemodynamics and to determine plaque morphology over the heart cycle, an accurate spatio-temporal model of the vessel under consideration is required. X-ray angiography and intravascular ultrasound (IVUS) represent the most commonly used diagnostic tools to assess coronary diseases. While conventional reconstructions from IVUS stack the frames as acquired during the catheter pullback to form a straight (pseudo-) 3-D volume, our comprehensive system considers the three-dimensional vessel curvature and reliably separates images originating from different heart phases. The fusion of the IVUS data with the pullback path as determined from x-ray angiography yields a geometrically accurate 4-D (3-D plus time) model of the coronary vasculature. This model is separated in a sequence of heart-phase-specific 3-D models that are used for calculating computational fluid dynamics (CFD). By determination of the inner and outer plaque boundaries, the plaque is subdivided into volume elements and mapped with the annotated finite element mesh generated during the CFD analysis. This allows a direct correlation of the local shear-stress data with plaque thickness in any given region of a vessel segment.