Many large-scale longitudinal imaging studies have been or are being widely conducted to better understand the progress of neurodegenerative and neuropsychiatric disorders or the normal brain development. Compared to cross-sectional imaging studies, the longitudinal imaging studies can identify subtle anatomical and functional changes and the causal role of time-dependent covariate (e.g., exposure) in disease process. However, analysis of longitudinal imaging data has been hindered by the lack of advanced image processing and statistical tools for analyzing complex and correlated imaging data along with behavioral and clinical data. Relatively, cross-sectional image processing and statistical tools have been developed and used, but they are in general not optimal in power. In response to PAR-06-411, the primary goal of this project is to develop new statistical tools and to evaluate these statistical tools and 4D image processing for analysis of imaging data, in combination with behavioral and clinical information obtained from longitudinal studies. As these tools are developed, they will be evaluated and refined through extensive Monte Carlo simulations and data analysis. Also, the efficacy of the tools developed under this grant will be tested by both simulated longitudinal datasets and the ADNI (Alzheimer's Disease Neuroimaging Initiative) dataset for early detection of Alzheimer's Disease (AD), respectively. Moreover, the companion software for all developed statistical tools, once validated, will be disseminated to imaging researchers through www.nitrc.org/, as we did for our brain image registration algorithm called HAMMER. This longitudinal analysis software will provide much needed imaging tools for analyzing complex, correlated imaging data in biomedical, behavioral, and social sciences. Thus, it is applicable to a variety of longitudinal neuroimaging studies, e.g., on major neurodegenerative diseases, neuropsychiatric disorders, substance use disorders, and brain development.
The project proposes to analyze imaging, behavioral, and clinical data from one large neuroimaging study on Alzheimer's diseases. New statistical methods are developed and applied to detect morphological differences of cortical and subcortical structures across time between Alzheimer patients and healthy subjects.
|Li, Tengfei; Xie, Fengchang; Feng, Xiangnan et al. (2018) Functional Linear Regression Models for Nonignorable Missing Scalar Responses. Stat Sin 28:1867-1886|
|Kong, Dehan; Ibrahim, Joseph G; Lee, Eunjee et al. (2018) FLCRM: Functional linear cox regression model. Biometrics 74:109-117|
|Zhang, Jingwen; Huang, Chao; Ibrahim, Joseph G et al. (2017) HFPRM: Hierarchical Functional Principal Regression Model for Diffusion Tensor Image Bundle Statistics. Inf Process Med Imaging 10265:478-489|
|Cornea, Emil; Zhu, Hongtu; Kim, Peter et al. (2017) Regression Models on Riemannian Symmetric Spaces. J R Stat Soc Series B Stat Methodol 79:463-482|
|Li, Jialiang; Huang, Chao; Zhu, Hongtu (2017) A Functional Varying-Coefficient Single-Index Model for Functional Response Data. J Am Stat Assoc 112:1169-1181|
|Pan, Wenliang; Wang, Xueqin; Wen, Canhong et al. (2017) Conditional local distance correlation for manifold-valued data. Inf Process Med Imaging 10265:41-52|
|Chow, Sy-Miin; Lu, Zhaohua; Sherwood, Andrew et al. (2016) Fitting Nonlinear Ordinary Differential Equation Models with Random Effects and Unknown Initial Conditions Using the Stochastic Approximation Expectation-Maximization (SAEM) Algorithm. Psychometrika 81:102-34|
|Hyun, Jung Won; Li, Yimei; Huang, Chao et al. (2016) STGP: Spatio-temporal Gaussian process models for longitudinal neuroimaging data. Neuroimage 134:550-562|
|Luo, Xinchao; Zhu, Lixing; Zhu, Hongtu (2016) Single-index varying coefficient model for functional responses. Biometrics 72:1275-1284|
|Huang, Meiyan; Nichols, Thomas; Huang, Chao et al. (2015) FVGWAS: Fast voxelwise genome wide association analysis of large-scale imaging genetic data. Neuroimage 118:613-27|
Showing the most recent 10 out of 59 publications