TY - JOUR
T1 - A statistical analysis of brain morphology using wild bootstrapping
AU - Zhu, Hongtu
AU - Ibrahim, Joseph G.
AU - Tang, Niansheng
AU - Rowe, Daniel B.
AU - Hao, Xuejun
AU - Bansal, Ravi
AU - Peterson, Bradley S.
N1 - Funding Information:
Manuscript received November 14, 2006; revised February 2, 2007. This work was supported in part by the Suzanne Crosby Murphy Endowment at Columbia University Medical Center and in part by the Thomas D. Klingenstein and Nancy D. Perlman Family Fund. The work of H. Zhu was supported by the NSF under Grant SES-0643663. The work of J. G. Ibrahim was supported by the NIH under Grant GM 70335 and Grant CA 74015. The work of N. Tang was supported by in part by NSFC under Grant 10561008 and in part by NSFYN under Grant 2004A0002M. The work of D. B. Rowe was supported by the NIH under Grant EB000215. The work of B. S. Peterson was supported in part by the NIDA under Grant DA017820 and in part by NIMH under Grant MH068318 and Grant K02-74677. Asterisk indicates corresponding author. *H. Zhu is with Department of Biostatistics and Biomedical Research Imaging Center, University of North Carolina, Chapel Hill, NC 27599-7420 USA (e-mail: hzhu@bios.unc.edu).
PY - 2007/7
Y1 - 2007/7
N2 - Methods for the analysis of brain morphology, including voxel-based morphology and surface-based morphometries, have been used to detect associations between brain structure and covariates of interest, such as diagnosis, severity of disease, age, IQ, and genotype. The statistical analysis of morphometric measures usually involves two statistical procedures: 1) invoking a statistical model at each voxel (or point) on the surface of the brain or brain subregion, followed by mapping test statistics (e.g., t test) or their associated p values at each of those voxels; 2) correction for the multiple statistical tests conducted across all voxels on the surface of the brain region under investigation. We propose the use of new statistical methods for each of these procedures. We first use a heteroscedastic linear model to test the associations between the morphological measures at each voxel on the surface of the specified subregion (e.g., cortical or subcortical surfaces) and the covariates of interest. Moreover, we develop a robust test procedure that is based on a resampling method, called wild bootstrapping. This procedure assesses the statistical significance of the associations between a measure of given brain structure and the covariates of interest. The value of this robust test procedure lies in its computationally simplicity and in its applicability to a wide range of imaging data, including data from both anatomical and functional magnetic resonance imaging (fMRI). Simulation studies demonstrate that this robust test procedure can accurately control the family-wise error rate. We demonstrate the application of this robust test procedure to the detection of statistically significant differences in the morphology of the hippocampus over time across gender groups in a large sample of healthy subjects.
AB - Methods for the analysis of brain morphology, including voxel-based morphology and surface-based morphometries, have been used to detect associations between brain structure and covariates of interest, such as diagnosis, severity of disease, age, IQ, and genotype. The statistical analysis of morphometric measures usually involves two statistical procedures: 1) invoking a statistical model at each voxel (or point) on the surface of the brain or brain subregion, followed by mapping test statistics (e.g., t test) or their associated p values at each of those voxels; 2) correction for the multiple statistical tests conducted across all voxels on the surface of the brain region under investigation. We propose the use of new statistical methods for each of these procedures. We first use a heteroscedastic linear model to test the associations between the morphological measures at each voxel on the surface of the specified subregion (e.g., cortical or subcortical surfaces) and the covariates of interest. Moreover, we develop a robust test procedure that is based on a resampling method, called wild bootstrapping. This procedure assesses the statistical significance of the associations between a measure of given brain structure and the covariates of interest. The value of this robust test procedure lies in its computationally simplicity and in its applicability to a wide range of imaging data, including data from both anatomical and functional magnetic resonance imaging (fMRI). Simulation studies demonstrate that this robust test procedure can accurately control the family-wise error rate. We demonstrate the application of this robust test procedure to the detection of statistically significant differences in the morphology of the hippocampus over time across gender groups in a large sample of healthy subjects.
KW - Heteroscedastic linear model
KW - Hippocampus
KW - Multiple hypothesis test
KW - Permutation test
KW - Robust test procedure
UR - http://www.scopus.com/inward/record.url?scp=34547301579&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=34547301579&partnerID=8YFLogxK
U2 - 10.1109/TMI.2007.897396
DO - 10.1109/TMI.2007.897396
M3 - Article
C2 - 17649909
AN - SCOPUS:34547301579
SN - 0278-0062
VL - 26
SP - 954
EP - 966
JO - IEEE Transactions on Medical Imaging
JF - IEEE Transactions on Medical Imaging
IS - 7
ER -