Abstract
This paper introduces a new framework for the segmentation of different brain structures (white matter, gray matter, and cerebrospinal fluid) from 3D MR brain images at different life stages. The proposed segmentation framework is based on a shape prior built using a subset of co-aligned training images that is adapted during the segmentation process based on first- and second-order visual appearance characteristics of MR images. These characteristics are described using voxel-wise image intensities and their spatial interaction features. To more accurately model the empirical grey level distribution of the brain signals, we use a linear combination of discrete Gaussians (LCDG) model having positive and negative components. To accurately account for the large inhomogeneity in infant MRIs, a higher-order Markov-Gibbs Random Field (MGRF) spatial interaction model that integrates third- and fourth- order families with a traditional second-order model is proposed. The proposed approach was tested and evaluated on 102 3D MR brain scans using three metrics: the Dice coefficient, the 95-percentile modified Hausdorff distance, and the absolute brain volume difference. Experimental results show better segmentation of MR brain images compared to current open source segmentation tools.
Highlights
Accurate delineation of brain tissues from magnetic resonance (MR) images is an essential step in human brain mapping and neuroscience [1,2,3]
This paper addresses brain segmentation from MR images at different life stages, having the infancy stage the most complicated one due to reduced contrast, higher noise [4], and inverse contrast between the white matter (WM) and gray matter (GM) [5], Fig 1
To overcome the aforementioned limitations, this paper proposes a novel technique for brain segmentation from MR images
Summary
Accurate delineation of brain tissues from magnetic resonance (MR) images is an essential step in human brain mapping and neuroscience [1,2,3]. Brain MRI segmentation faces challenges stemming from image noise, magnetic field inhomogeneities, artifacts such as partial volume effects, and discontinuities of boundaries due to similar visual appearance of adjacent brain structures. This paper addresses brain segmentation from MR images at different life stages, having the infancy stage the most complicated one due to reduced contrast, higher noise [4], and inverse contrast between the white matter (WM) and gray matter (GM) [5], Fig 1. Segmentation of the brain at later stages might be relatively easier, as the contrast between different types of tissue is much better, and the signal-to-noise ratio (SNR) are improved, Fig 1.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
