Dissertations, Theses, and Capstone Projects

Date of Degree

2-2025

Document Type

Master's Thesis

Degree Name

Master of Science

Program

Cognitive Neuroscience

Advisor

Junghoon J. Kim

Subject Categories

Cognitive Neuroscience | Nervous System | Neurology | Neurosciences | Social and Behavioral Sciences | Trauma

Keywords

Traumatic brain injury, Thalamus, Atrophy, Neurodegeneration, MRI

Abstract

Traumatic brain injury (TBI) is a significant public health concern and often leads to widespread brain atrophy in the months to years following a moderate-to-severe TBI. The thalamus is particularly vulnerable to the mechanical forces of TBI and frequently shows a higher rate of atrophy compared to other gray matter structures. While whole brain morphometry studies have reported global and regional thalamic atrophy in moderate-to-severe TBI, variability in time postinjury and a lack of standardized longitudinal assessments have made it difficult to determine the exact progression of thalamic atrophy. Our group previously identified a unique temporal pattern of global thalamic atrophy, where the thalamus as a whole displayed both cross-sectional and longitudinal atrophy at three specific post-injury time points. This led us to further investigate the spatiotemporal profile of thalamic atrophy using voxel-wise tensor-based morphometry, allowing for a more precise analysis of the morphological changes within the structure. Structural T1- weighted MPRAGE scans were acquired from 37 moderate-to-severe TBI patients at three standardized post-injury time points (3, 6, and 12 months), and 33 demographically matched controls were scanned once. Tensor-based morphometry was used to quantify cross-sectional and longitudinal thalamic atrophy. ANTs SyN registration was used to warp each subject to their singlesubject template and then a group template. Jacobian determinant maps were calculated to quantify voxel-wise volume changes. Group differences between controls and TBI patients at 3 months postinjury were analyzed by performing a nonparametric two-sample t-test via FMRIB Software Library (FSL) Randomise, with age and total intracranial volume as covariates. Longitudinal changes between 3 and 12 months post-injury were assessed by performing repeated measures regression via FSL Randomise, with time from injury as a predictor for thalamic atrophy, and age and total intracranial volume as covariates. We found widespread thalamic atrophy at 3 months, with peaks in the bilateral medial dorsal nucleus (MD). Longitudinal atrophy between 3 and 12 months was observed in several subregions, with peak atrophy in the bilateral pulvinar (PV) and right anterior nucleus (ANT). An exploratory neurobehavioral analysis revealed a significant relationship between peak thalamic atrophy in the MD and clinical/neuropsychological outcomes. Finally, functional connectivity profiles associated with regions of peak thalamic atrophy were identified using Neurosynth.org.

Download

This work is embargoed and will be available for download on Monday, February 01, 2027

Share

COinS