A fuzzy rules-based segmentation method for medical images analysis

Abstract

Medical imaging mainly manages and processes missing, ambiguous, complementary, redundant and distorted data and information has a strong structural character. This paper reports a new (semi)automated and supervised method for the segmentation of brain structures using a rule-based fuzzy system. In the field of biomedical image analysis fuzzy logic acts as a unified framework for representing and processing both numerical and symbolic information, as well as structural information constituted mainly by spatial relationships. The developed application is for the segmentation of brain structures in CT (computer tomography) images. Promising results show the superiority of this knowledge-based approach over best traditional techniques in terms of segmentation errors. The quantitative assessment of this method is made by comparing manually and automatic segmented brain structures by using some indexes evaluating the accuracy of contour detection and spatial location. Though the proposed methodology has been implemented and successfully used for modeldriven in medical imaging, it is general enough and may be applied to any imagistic object that can be expressed by expert knowledge and morphological images. © 2006-2013 by CCC Publications.