TY - GEN
T1 - Multivariate tensor-based morphometry on surfaces
T2 - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009
AU - Wang, Yalin
AU - Toga, Arthur W.
AU - Thompson, Paul M.
AU - Chan, Tony F.
AU - Zhang, Jie
PY - 2009
Y1 - 2009
N2 - We apply multivariate tensor-based morphometry to study lateral ventricular surface abnormalities associated with HIV/AIDS. We use holomorphic one-forms to obtain a conformal parameterization of ventricular geometry, and to register lateral ventricular surfaces across subjects. In a new development, we computed new statistics on the Riemannian surface metric tensors that encode the full information in the deformation tensor fields. We applied this framework to 3D brain MRI data, to map the profile of lateral ventricular surface abnormalities in HIV/AIDS (11 subjects). Experimental results demonstrated that our method powerfully detected brain surface abnormalities. Multivariate Hotelling's T 2 statistics on the local Riemannian metric tensors, computed in a log-Euclidean framework, detected group differences with greater power than other surface-based statistics including the Jacobian determinant, largest and least eigenvalue, or the pair of eigenvalues of the Jacobian matrix. Computational anatomy studies may therefore benefit from surface parameterization using differential forms and tensor-based morphometry, in the log-Euclidean domain, on the resulting surface tensors.
AB - We apply multivariate tensor-based morphometry to study lateral ventricular surface abnormalities associated with HIV/AIDS. We use holomorphic one-forms to obtain a conformal parameterization of ventricular geometry, and to register lateral ventricular surfaces across subjects. In a new development, we computed new statistics on the Riemannian surface metric tensors that encode the full information in the deformation tensor fields. We applied this framework to 3D brain MRI data, to map the profile of lateral ventricular surface abnormalities in HIV/AIDS (11 subjects). Experimental results demonstrated that our method powerfully detected brain surface abnormalities. Multivariate Hotelling's T 2 statistics on the local Riemannian metric tensors, computed in a log-Euclidean framework, detected group differences with greater power than other surface-based statistics including the Jacobian determinant, largest and least eigenvalue, or the pair of eigenvalues of the Jacobian matrix. Computational anatomy studies may therefore benefit from surface parameterization using differential forms and tensor-based morphometry, in the log-Euclidean domain, on the resulting surface tensors.
KW - Holomorphic one-form
KW - Multivariate tensor-based morphometry
KW - Surface modeling
UR - http://www.scopus.com/inward/record.url?scp=70449379857&partnerID=8YFLogxK
U2 - 10.1109/ISBI.2009.5193000
DO - 10.1109/ISBI.2009.5193000
M3 - Conference contribution
AN - SCOPUS:70449379857
SN - 9781424439324
T3 - Proceedings - 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009
SP - 129
EP - 132
BT - Proceedings - 2009 IEEE International Symposium on Biomedical Imaging
Y2 - 28 June 2009 through 1 July 2009
ER -
