A meta-algorithm for brain extraction in MRI

Laboratory of Neuro Imaging, Department of Neurology, David Geffen School of Medicine at UCLA, 710 Westwood Plaza, Los Angeles, CA 90095-1769, USA.
NeuroImage (Impact Factor: 6.36). 11/2004; 23(2):625-37. DOI: 10.1016/j.neuroimage.2004.06.019
Source: PubMed


Accurate identification of brain tissue and cerebrospinal fluid (CSF) in a whole-head MRI is a critical first step in many neuroimaging studies. Automating this procedure can eliminate intra- and interrater variance and greatly increase throughput for a labor-intensive step. Many available procedures perform differently across anatomy and under different acquisition protocols. We developed the Brain Extraction Meta-Algorithm (BEMA) to address these concerns. It executes many extraction algorithms and a registration procedure in parallel to combine the results in an intelligent fashion and obtain improved results over any of the individual algorithms. Using an atlas space, BEMA performs a voxelwise analysis of training data to determine the optimal Boolean combination of extraction algorithms to produce the most accurate result for a given voxel. This allows the provided extractors to be used differentially across anatomy, increasing both the accuracy and robustness of the procedure. We tested BEMA using modified forms of BrainSuite's Brain Surface Extractor (BSE), FSL's Brain Extraction Tool (BET), AFNI's 3dIntracranial, and FreeSurfer's MRI Watershed as well as FSL's FLIRT for the registration procedure. Training was performed on T1-weighted scans of 136 subjects from five separate data sets with different acquisition parameters on separate scanners. Testing was performed on 135 separate subjects from the same data sets. BEMA outperformed the individual algorithms, as well as interrater results from a subset of the scans, when compared for the mean Dice coefficient, a rating of the similarity of output masks to the manually defined gold standards.

Full-text preview

Available from:
  • Source
    • "Beare et al. [16] introduced markerbased watershed scalper (MBWSS) for brain extraction in T1-weighted MR images that is built using filtering and segmentation components from the Insight Toolkit (ITK) framework. Rex et al. [17] developed a meta-algorithm that uses four freely available brain extraction algorithms: brain surface extractor (BSE) [18], brain extraction tool (BET) [4], 3dIntracranial [19], and MRI watershed from FreeSurfer [20]. For extracting the brain, an atlas is used to define which extraction algorithm or combination of extractors works best defining the brain in each anatomic region. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we propose a novel framework for the automated extraction of the brain from T1-weighted MR images. The proposed approach is primarily based on the integration of a stochastic model (a two-level Markov-Gibbs random field (MGRF)) that serves to learn the visual appearance of the brain texture, and a geometric model (the brain iso-surfaces) that preserves the brain geometry during the extraction process. The proposed framework consists of three main steps: (i) Following bias correction of the brain, a new 3D MGRF having a 26- pairwise interaction model is applied to enhance the homogeneity of MR images and preserve the 3D edges between different brain tissues; (ii) The non-brain tissue found in the MR images is initially removed using the brain extraction tool (BET), and then the brain is parceled to nested iso-surfaces using a fast marching level set method; (iii) Finally, a classification step is applied in order to accurately remove the remaining parts of the skull without distorting the brain geometry. The classification of each voxel found on the iso-surfaces is made based on the first- and second-order visual appearance features. The first-order visual appearance is estimated using a linear combination of discrete Gaussians (LCDG) to model the intensity distribution of the brain signals. The second-order visual appearance is constructed using an MGRF model with analytically estimated parameters. The fusion of the LCDG and MGRF, along with their analytical estimation, allows the approach to be fast and accurate for use in clinical applications. The proposed approach was tested on in-vivo data using 300 infant 3D MR brain scans, which were qualitatively validated by an MR expert. In addition, it was quantitatively validated using 30 data sets based on three metrics: the Dice coefficient, the 95-percentile modified Hausdorff distance, and absolute brain volume difference. Results showed the capability of the proposed approach, outperforming four widely used brain extraction tools: BET, BET2, brain surface extractor (BSE), and infant brain extraction and analysis toolbox (iBEAT). Experiments conducted also proved that the proposed framework can be generalized to adult brain extraction as well.
    Full-text · Article · Mar 2015
    • "Segonne et al. [8] presented a hybrid approach that combined watershed algorithms and deformable surface models. Rex et al. [9] developed a meta-algorithm that executes many brain extraction algorithms and a registration procedure followed by an approach to combine the results. In summary, different brain extraction approaches have been developed; however, the existing approaches have their own drawbacks. "
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a novel approach for extracting the brain from 3D T1-weighted MR images. The proposed approach combines a stochastic two-level Markov-Gibbs random field (MGRF) image model with a geometric model that parcels the brain into a set of nested iso-surfaces using a fast marching level setmethod. The classification of each brain voxel found on the iso-surfaces is performed based on the first-order (a linear combination of discrete gaussian (LCDG) model) and second-order (an MGRF model with analytically estimated parameters) visual appearance features of the brain structures. Our approach is tested on 280 infant 3D MR brain scans and evaluated on 9 data sets using the Dice coefficient, the 95-percentile modified Hausdorff distance, and absolute brain volume difference. Experimental results showed that the fusion of the stochastic and geometric models of brain MRI data has led to more accurate brain extraction, when compared with other widely-used brain extraction tools, such as BET, BET2, and brain surface extractor (BSE).
    No preview · Article · Jan 2015
  • Source
    • "In meta-algorithm (Rex et al., 2004; Shi et al., 2012), several existing brain extraction methods are combined to compensate the weaknesses of each method. However, the model should be specifically designed through meta-algorithm to gain optimum performance when new data are sufficiently different from previous training datasets (Rex et al., 2004). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Brain extraction is an important procedure in brain image analysis. Although numerous brain extraction methods have been presented, enhancing brain extraction methods remain challenging because brain MRI images exhibit complex characteristics, such as anatomical variability and intensity differences across different sequences and scanners. To address this problem, we present a Locally Linear Representation-based Classification (LLRC) method for brain extraction. A novel classification framework is derived by introducing the locally linear representation to the classical classification model. Under this classification framework, a common label fusion approach can be considered as a special case and thoroughly interpreted. Locality is important to calculate fusion weights for LLRC; this factor is also considered to determine local anchor embedding is more applicable in solving locally linear coefficients compared with other linear representation approaches. Moreover, LLRC supplies a way to learn the optimal classification scores of the training samples in the dictionary to obtain accurate classification. The International Consortium for Brain Mapping and the Alzheimer's Disease Neuroimaging Initiative databases were used to build a training dataset containing 70 scans. To evaluate the proposed method, we used three public available datasets (IBSR1, IBSR2, and LPBA40, with a total of 78 scans). Experimental results demonstrate that the proposed method outperforms the four common brain extraction methods (BET, BSE, GCUT, and ROBEX), and is comparable to the performance of BEaST, while being more accurate on some datasets compared with BEaST.
    Full-text · Article · May 2014 · NeuroImage
Show more