Measuring and Segmentation in CT Data using Deformable Models

Krajíček,V., Pelikán,J., Horák,M.

Abstract:
Accurate measuring of physical properties of human body has great importance for determining the best treatment. Our work aims at measuring volumes of organs such as kidney or liver in image data obtained from computed tomography (CT). We take advantage of long-time research in the area of deformable models. We have developed parametric model using closed B-Spline curves and have formulated energetic equation for their iterative evolution. Interior and exterior intensity distributions are taken into account, together with upcoming shape and position of regions in neighbouring slices of multi-slice CT data. This approach does not require gradient information, which is unreliable in medical images.