Multiscale Adaptive Marginal Analysis of Longitudinal Neuroimaging Data with Time‐Varying Covariates

Abstract

Summary Neuroimaging data collected at repeated occasions are gaining increasing attention in the neuroimaging community due to their potential in answering questions regarding brain development, aging, and neurodegeneration. These datasets are large and complicated, characterized by the intricate spatial dependence structure of each response image, multiple response images per subject, and covariates that may vary with time. We propose a multiscale adaptive generalized method of moments (MA-GMM) approach to estimate marginal regression models for imaging datasets that contain time-varying, spatially-related responses and some time-varying covariates. Our method categorizes covariates into types to determine the valid moment conditions to combine during estimation. Further, instead of assuming independence of voxels (the components that make up each subject’s response image at each time point) as many current neuroimaging analysis techniques do, this method “adaptively smoothes” neuroimaging response data, computing parameter estimates by iteratively building spheres around each voxel and combining observations within the spheres with weights. MA-GMM’s development adds to the few available modeling approaches intended for longitudinal imaging data analysis. Simulation studies and an analysis of a real longitudinal imaging dataset from the Alzheimer’s Disease Neuroimaging Initiative are used to assess the performance of MA-GMM.

2014 Update of the Alzheimer's Disease Neuroimaging Initiative: A review of papers published since its inception

“…Significant group differences were found between control subjects, MCI patients who did or did not subsequently convert to AD (MCI-c and MCI-nc, respectively) and AD patients, and the authors suggested that the method may be particularly useful as an AD biomarker in conjunction with shape analysis, as both approaches leverage complementary information. Two new approaches for analyzing longitudinal MRI data sets were reported by Skup et al [380] and Bernal-Rusiel et al [381]. The multiscale adaptive generalized method of moments [380] tackles the problem of analysis of longitudinal MRI data sets that have multiple response images per subject.…”

Genome-wide association analysis of secondary imaging phenotypes from the Alzheimer's disease neuroimaging initiative study

“…Although we have designed the simulation studies by assuming a single quantitative secondary trait and the additive mode of inheritance, the studies can be easily extended to consider other kind of secondary traits such as binary traits (Wang and Shete (2011); Chen et al (2013)), longitudinal traits (Skup et al (2012); Xu et al (2014)), multiple traits (Lin et al (2012); Zhang et al (2014); Zhu et al (2014)) as well as other modes of inheritance. We only include AD patients and CN subjects in the GWAS of ADNI data in the paper, but we may include the MCI subjects in our GWAS and treat them as controls by following the analysis in Kim and Pan (2015).…”

FVGWAS: Fast voxelwise genome wide association analysis of large-scale imaging genetic data

“…Although it is common to use Gaussian smoothing with a prefixed bandwidth, it may introduce substantial bias in the statistical results (Li et al, 2012, 2013). Although several multi-scale adaptive regression models (MARMs) have been developed for the group analysis of imaging data (Li et al, 2011; Skup et al, 2012; Li et al, 2012; Polzehl et al, 2010), these methods are not computationally feasible even for thousands of SNPs. It is critically important to develop some novel methods to explicitly incorporate the spatial feature of the imaging data in FVGWAS, while achieving computational efficiency for ultra-high-resolution imaging and whole-genome sequencing.…”