Automatic contour extraction from HVEM images of microvessels

This work is motivated by the problem raised in the biomedical research utilizing high voltage electron microscope (HVEM) to study microvessels. There, the contour data of cross-sections of microvessels are fed into a special electronic device to reconstruct three-dimensional pictures of microvessels. Useful parameters characterizing the microvessels can then be derived. However, the contour data acquisition is routinely conducted manually. This is both tedious and time-consuming. To our knowledge there is no commercially available software that can avoid this cumbersome manual procedure. In this paper, a synthetic method, consisting of several, computer image processing techniques, is presented and tested on the HVEM images of thick cross sections of microvessels. Besides the application of several conventional digital image processing techniques some particular problem-oriented techniques are developed in the course of solving the problem. The results demonstrate that it is possible to use computer image processing techniques instead of manual methods to obtain contour data from electromicroscopic pictures of the kind described above.