A meta-algorithm for brain extraction in MRI. Export

@article{citeulike:939983, abstract = {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.}, address = {Laboratory of Neuro Imaging, Department of Neurology, David Geffen School of Medicine at UCLA, 710 Westwood Plaza, Los Angeles, CA 90095-1769, USA.}, author = {Rex, D. E. and Shattuck, D. W. and Woods, R. P. and Narr, K. L. and Luders, E. and Rehm, K. and Stolzner, S. E. and Rottenberg, D. A. and Toga, A. W.}, citeulike-article-id = {939983}, citeulike-linkout-0 = {http://dx.doi.org/10.1016/j.neuroimage.2004.06.019}, citeulike-linkout-1 = {http://view.ncbi.nlm.nih.gov/pubmed/15488412}, citeulike-linkout-2 = {http://www.hubmed.org/display.cgi?uids=15488412}, doi = {10.1016/j.neuroimage.2004.06.019}, issn = {1053-8119}, journal = {Neuroimage}, keywords = {bema, csf}, month = {October}, number = {2}, pages = {625--637}, posted-at = {2008-11-17 10:38:34}, priority = {2}, title = {A meta-algorithm for brain extraction in MRI.}, url = {http://dx.doi.org/10.1016/j.neuroimage.2004.06.019}, volume = {23}, year = {2004} }

TY - JOUR ID - citeulike:939983 L3 - citeulike-article-id:939983 N2 - 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. IS - 2 JF - Neuroimage SN - 1053-8119 AD - Laboratory of Neuro Imaging, Department of Neurology, David Geffen School of Medicine at UCLA, 710 Westwood Plaza, Los Angeles, CA 90095-1769, USA. EP - 637 TI - A meta-algorithm for brain extraction in MRI. VL - 23 SP - 625 KW - bema KW - csf AU - Rex, DE AU - Shattuck, DW AU - Woods, RP AU - Narr, KL AU - Luders, E AU - Rehm, K AU - Stolzner, SE AU - Rottenberg, DA AU - Toga, AW PY - 2004/10// UR - http://dx.doi.org/10.1016/j.neuroimage.2004.06.019 ER -