Fuzziness and PDE based models for the segmentation of medical images

Abstract

This paper presents a generic strategy to facilitate the segmentation of anatomical structures in medical images the segmentation of anatomical structures in medical images. The segmentation is performed using an adapted partial differential equation (PDE) based model by fuzzy c-means classification. The PDE defines a geometrical active contour model, which also uses the fuzzy decision to converge into the final contour. Furthermore, the fuzzy reasoning exploits a priori statistical information from several knowledge sources based on histogram analysis and the intensity values of the anatomical structures under consideration. The advantage of using a geometric representation is that the algorithm can handle changes in the topology of the shape as the curve/surface evolves in time. Moreover, the fuzzy-PDE model is able to solve some drawbacks present at other PDE based models.