Determination of three-dimensional flow velocity distributions from single-plane angiograms

Velocity data is important for assessment and study of diseased vessels. Techniques for in vivo velocity measurements remain limited. Understanding 3D flow velocity fields can be valuable during interventional procedures. In this work, we present a new method of measuring 3D velocity data from single-view angiographic sequence.;We have generated virtual angiographic sequences to test our technique. Sequences are developed using a simplified flow field. Contrast material is propagated through the flow field. A series of ray cast images are generated to simulate single-view radiographic projections of the contrast leading edge as it travels through the region of interest. Images are produced without noise and with a signal to noise ratio (SNR) of 10. We have developed a method to reconstruct a 3D distribution of contrast material from a single-view angiogram using bilateral symmetry. We take advantage of x-ray attenuation and projection ratios to obtain our 3D reconstruction. We present a method to generate velocity data from time-sampled contrast-interface surfaces. To do this, we establish corresponding points on the surfaces using assumed streamlines. Streamlines are generated based on the enclosed geometry of the vessel.;Geometry reconstruction errors are within a pixel (noiseless) and less than 5% (volumetric, SNR=10). Average velocity errors between surfaces are range from 10% and 20% (noiseless) and 20% for select SNR=10 cases. The reconstructed flow field error is 25% in regions away from the vessel wall.;In this thesis we describe a framework used to obtain 3D velocity data using ubiquitous procedures and equipment.