Fractal Analysis for Symmetry Plane Detection in Neuroimages

Abstract

Despite the considerable amount of research, brain symmetry plane detection is still an open problem. In this paper, we present a novel method for symmetry plane detection in magnetic resonance (MR) neuroimages based on the textural information and underlying brain's physiological structure. Fractal dimension and lacunarity analysis are used to locate the symmetry plane of the brain. The method was tested on MR data while analyzing the robustness against intensity non-uniformity, noise, and pathology. The proposed method does not need skull-stripping like pre-processing of MR images. The method was compared with another commonly used technique. The results were evaluated by an expert. The experimental results show the viability of our approach.