Fully Automatic Method for 3D T1-Weighted Brain Magnetic Resonance Images Segmentation

Accurate segmentation of brain MR images is of interest for many brain disorders. However, dueto several factors such noise, imaging artefacts, intrinsic tissue variation and partial volumeeffects, brain extraction and tissue segmentation remains a challenging task. So, in this paper, afull automatic method for segmentation of anatomical 3D brain MR images is proposed. Themethod consists of many steps. First, noise reduction by median filtering is done; secondsegmentation of brain/non-brain tissue is performed by using a Threshold Morphologic BrainExtraction method (TMBE). Then initial centroids estimation by gray level histogram analysis isexecuted, this stage yield to a Modified version of Fuzzy C-means Algorithm (MFCM) that is usedfor MRI tissue segmentation. Finally 3D visualisation of the three clusters (CSF, GM and WM) isperformed. The efficiency of the proposed method is demonstrated by extensive segmentationexperiments using simulated and real MR images. A confrontation of the method with similarmethods of the literature has been undertaken trough different performance measures. TheMFCM for tissue segmentation introduce a gain in rapidity of convergence of about 70%.