Development and application of a system for 3-D measurement of biostructures by computational geometry

点击翻译按钮获取中文摘要

Although the pathological studies deal with 3-dimensional biostructures such as the organs, tissues and cells of man and animals, the actural means of observation is confined to 2-dimensional sectional pictures, The tchniques of stereology, which is based on stochastic geometty, cannot work in most aspects of pathology where one is often unable to assimilate actual biostructures with pertinent geometric models. In view of this, we designed a system by which to perform 3-dimensional measurement of biostructures by computational geometry from structural data in putted from serial sections. We managed to establish algorithms for the following three tasks, in which, the system is ready to application to actual studies.1. Computer-assisted granulometry : This is to obtain information on the number and size distribution of granular biostructures such as the cells, their nuclei or the islets of Langerhans. The computer scans through "sampling bricks, " a 3-dimensional sampling probe constructed in a tissue space containing granules and inputted from serial sections.2. Expression of vascular pattern with the spatial distance distribution : An algorithm was devised to compare the pattern of 3-dimensional architecture with the distribution of distances from randomly positioned points to the structure of interest. The method was applied to the analysis of vasculature in livers with chronic hepatitis and cirrhosis, to quantitatively describe how the vascular arrangement deviates from the norm with the progress of cirrhogenesis.3. Three-dimensional topological analysis : There are features of diseased organs that can be analyzed only in topological terms, such as the form of carchinoma cell nests. This requirement was met by developing an algorithm for computer-aided measurement of topological invariable (1st Betti number). We aere now finding a solution for the three-dimensional thinning of biostructures, which will give us a method to visualize their skeleton.