Related Experiment Video
Updated: May 10, 2026
Translational Brain Mapping at the University of Rochester Medical Center: Preserving the Mind Through Personalized Brain Mapping
Published on: August 12, 2019
Intrinsic Regression Models for Medial Representation of Subcortical Structures
Xiaoyan Shi1, Hongtu Zhu, Joseph G Ibrahim
1H. Zhu is Professor of Biostatistics ( hzhu@bios.unc.edu ), J. G. Ibrahim is Alumni Distinguished Professor of Biostatistics ( ibrahim@bios.unc.edu ), and X. Shi was Ph.d student ( amy.shi@sas.com ), Department of Biostatistics and Biomedical Research Imaging Center, University of North Carolina at Chapel Hill, NC 27599-7420. F. Liang is Professor of Statistics ( fliang@stat.tamu.edu ), Department of Statistics, Texas A & M University, College Station, TX 77843-3143. Jeffrey Lieberman is Lawrence C. Kolb Professor of Psychiatry ( jlieberman@pi.cpmc.Columbia.edu ), Department of Psychiatry, Columbia University Medical Center, 1051 Riverside Drive, New York, New York 10032, U.S.A. M. Styner is Assistant Professor ( yasheng.chen@med.unc.edu ), Department of Computer Science and Psychiatry, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599.
This study introduces a new statistical model for analyzing brain structure shapes, particularly the hippocampus, in individuals with schizophrenia. The model helps identify morphological differences linked to diagnostic status.
Area of Science:
- Statistics
- Neuroimaging
- Biometry
Background:
- Subcortical structures exhibit variability best described on Riemannian manifolds.
- Understanding morphological changes in conditions like schizophrenia requires advanced statistical modeling.
Purpose of the Study:
- To develop a semiparametric model for subcortical structure variability on Riemannian manifolds.
- To associate these shape variations with covariates like diagnostic status, age, and gender.
- To apply the model to detect hippocampal differences in schizophrenia patients.
Main Methods:
- A two-stage estimation procedure involving an intrinsic least squares estimator and an annealing evolutionary stochastic approximation Monte Carlo algorithm.
- Development of estimating equations for efficient parameter estimation.
- Utilizing Wald statistics for hypothesis testing and establishing limiting distributions.
- Simulation studies to assess parameter estimate accuracy and statistical test performance.
Main Results:
- The developed semiparametric model effectively describes subcortical shape variability.
- The two-stage estimation procedure provides accurate parameter estimates.
- Wald statistics demonstrate reliable performance in hypothesis testing.
- The method successfully detected morphological differences in hippocampi between schizophrenia patients and controls.
Conclusions:
- The proposed semiparametric model and estimation techniques are effective for analyzing complex shape data in neuroimaging.
- This approach can reveal significant morphological differences in subcortical structures related to psychiatric conditions.
- The findings contribute to a better understanding of brain morphology in schizophrenia.
