Statistical Evaluation of Velocity Estimations from Temporal Fluoroscopic Angiography

The vascular diseases are reflected by disorders in morphology and hemodynamics. For the diagnosis and treatment of these diseases, medical imagistic investigations are needed. They explore the morphological and functional aspects of blood circulation. The extraction of the blood flow from daily clinical acquisitions of X-ray angiography is of great importance for the physicians. The paper presents the implementation of bolus tracking algorithm which is related to contrast agent propagation into the blood stream for coronary vessels. The time density curve (TDC) is computed because it is a temporal evolution of the grey levels of a group of pixels tagged into an image set. For each image two TDCs are plotted into a dedicated MATLAB® based graphical user interface (GUI). The medical imagistic expert manually selects the proximal and the distal regions which delineates the vessel segment to be analyzed. The vessel length between the two regions is estimated enabling the velocity estimation. The image set is acquired from 20 patients, and it is investigated through 11 different transit time estimation methods. The results conduct to a future prototype system for diagnosis of vascular hemodynamic disorders.