Contour Processing and 3D Visualization in Medical Images
Katrina Boločko, Aleksandrs Glazs · 2010
Introduction. One of the most important tasks in medical image processing is pathology zone extraction and its analysis. In those cases, when a surgical intervention is required, 3D visualization of the medical image aids physicians in planning the operation. In order to obtain a 3D model of the medical object and its surroundings, the desired medical object must be first extracted from the medical image slices. Then medical objects contour control points, which are further used for 3D visualization of the object, must be selected. In this paper, a method of visualizing a selected medical object is proposed. The method uses a polygonal-based 3D model. The necessary control points for the 3D model are acquired from the extracted contour and a 3D mesh is constructed and then visualized in the developed software. Proposed method. The proposed approach combines two methods – contour processing for obtaining control points and visualization – the construction of a 3D surface from the acquired points. Firstly, a medical object in the image is selected and its contour processed to obtain control points that are necessary to reconstruct the 3D model. The proposed method for contour processing is based on the modification of a method described in [1]. This modification includes combining the method with the clockwise point selection [2] and implementing an adaptive threshold Tij for each part between the adjacent points pi,j and pi,j+1 of the adaptive contour on slice i, in a way that a lesser distance would require a stricter threshold: