Segmentation of Cerebrospinal Fluid and Internal Brain Nuclei in Brain Magnetic Resonance Images

Abstract

Brain tissue segmentation on structural Magnetic Resonance Imaging (MRI) has received considerable attention. Quantitative analysis of MR images of the brain is of interest in order to study the aging brain in epidemiological studies, to better understand how diseases affect the brain and to support diagnosis in clinical practice. Manual quantitative analysis of brain imaging data is a tedious and time-consuming procedure, prone to observer variability. Therefore, there is a large interest in automatic analysis of MR brain imaging data, especially segmentation of Cerebrospinal Fluid (CSF), Gray Matter (GM) and White Matter (WM). This paper presents a fully automated method for the segmentation of cerebrospinal fluid and internal brain nuclei from T1-weighted MRI head scans. The proposed methodology performs intensity based thresholding to get the boundaries between gray matter, white matter, cerebrospinal fluid and others. Combined with preprocessing techniques and incorporating mathematical morphology, we first perform the extraction of brain cortex. Subsequently, the cerebrospinal fluid is segmented by using orthogonal polynomial transform. Finally, the gray matter and the white matter regions in the MRI are segmented based on the intensity values. Experimental results show that the proposed method achieves reasonably good segmentation. The comparative analysis depicts that the proposed methodology shows better segmentation results with some other existing techniques like FAST, SPM5, k-nearest neighbor (k-NN) classifier, and a conventional k-NN.