Two-stage empirical likelihood for longitudinal neuroimaging data

TL;DR

This paper introduces a two-stage empirical likelihood method for analyzing longitudinal neuroimaging data, effectively handling spatial dependence and classifying covariate types without modeling temporal correlation.

Contribution

It develops a novel two-stage adjusted exponentially tilted empirical likelihood approach for spatial analysis of longitudinal neuroimaging data, accommodating spatial dependence and different covariate types.

Findings

Method performs well in simulation studies.

Successfully applied to detect hippocampal changes in schizophrenia.

Identifies differences between patients and healthy subjects.

Abstract

Longitudinal imaging studies are essential to understanding the neural development of neuropsychiatric disorders, substance use disorders, and the normal brain. The main objective of this paper is to develop a two-stage adjusted exponentially tilted empirical likelihood (TETEL) for the spatial analysis of neuroimaging data from longitudinal studies. The TETEL method as a frequentist approach allows us to efficiently analyze longitudinal data without modeling temporal correlation and to classify different time-dependent covariate types. To account for spatial dependence, the TETEL method developed here specifically combines all the data in the closest neighborhood of each voxel (or pixel) on a 3-dimensional (3D) volume (or 2D surface) with appropriate weights to calculate adaptive parameter estimates and adaptive test statistics. Simulation studies are used to examine the finite sample performance of the adjusted exponential tilted likelihood ratio statistic and TETEL. We demonstrate the application of our statistical methods to the detection of the difference in the morphological changes of the hippocampus across time between schizophrenia patients and healthy subjects in a longitudinal schizophrenia study.