Vessel Enhancing Diffusion Filter

Abstract

This paper describes vessel enhancing diffusion (VED) filters implemented using the Insight Toolkit (ITK). The filters are implementation of the VED algorithm developed by Manniesing et al . The VED algorithm follows a multiscale approach to enhance vessels using anisotropic diffusion scheme guided by vesselness measure at a pixel level. Vesselness is determined by geometrical analysis of the Eigen system of the Hessian matrix. For this purpose, a smoothed version of the Frangi's vesselness function is formulated. Experiments were conducted to evaluate the performance of the VED filters in enhancing vessels in lung CT scan.