An Automatic Segmentation Method for MRI Multiparametric Volumes

Abstract

Abstract-Purpose of this work is the design and implementation of an automated method for digital volume segmentation, based on multi-parametric densities, fuzzy topology, and adaptive growth mechanism. The processing objective is the global segmentation of the digital volume, that is its partitioning into significant connected subsets, in a fully automatic way. The main advantage consists in the very nature of the algorithm that enables the automatic segmentation by running an iterative process that adapts to the volume at hand and does not require any user intervention. The designed method can be applied to multi-parametric volumes where different characteristics are available to analyze the same target. The robustness of the method has been evaluated and verified through statistical parameters, that will be discussed below, after application on volumes of biomedical images obtained through Magnetic Resonance Imaging. IndexTerms-Segmentation, fuzzy processing, connectedness, multi-parametric data fusion.