Calibración geométrica de un equipo de angiografía digital

Abstract

Digital X-ray angiography is a medical imaging modality that enables acquisition of high quality images of coronary vessels and heart cavities. The angiography images are used mainly for quantitative assessment of coronary obstructions. However, in clinical practice, the measurement of the lesion length is obtained by comparing the size of the lesion with respect to catheter width. These manual techniques add imprecision to the lesion measurements since they do not consider the acquisition geometry of the X-ray system. Furthermore, three-dimensional reconstruction techniques have emerged as a solution to obtain more reliable lesion length measurements. However, these techniques require the information about the X-ray equipment acquisition geometry in order to generate reliable 3D models. This article proposes a computational method to obtain intrinsic parameters of the Xray angiography system, taking into account the image radial distortion. The goal of this method is the estimation of reliable calibration parameters in order to obtain the computational 3D model of the arteries. The calibration method is evaluated by estimating the distance, measured in pixels, between detected image points and projected ones. © Springer-Verlag Berlin Heidelberg 2007.