MRI-based Morphometry During Neurodevelopment PDF Download

Author: Alonso Cárdenas de la Parra
Publisher:
ISBN:
Category :
Languages : en
Pages :

Get Book Here

Book Description
"Magnetic resonance imaging (MRI) is a non-invasive imaging technique commonly used to study neurological changes associated with disease. Morphometric analysis of MRI focuses on the quantitative analysis of size and shape of regions within the brain, thus providing a powerful tool for observing neurological changes. While many image processing tools have been developed for morphometric analyses, most have been designed with the adult population in mind. However, numerous diseases and disorders present themselves during childhood, and pediatric neuroimaging studies can help us better characterize them, aiding in understanding their etiology and progression, and providing information for better diagnostic tools. Pediatric neuroimaging studies present a series of unique challenges due to the normal neurodevelopment occurring during childhood and adolescence, with the brain undergoing substantial changes in size, shape, and maturation. The primary objective of this thesis is to design, test and validate morphometric techniques for pediatric neuroimaging studies. The first proposed technique focuses on early childhood (6-24 months of age), and consists of the application of tensor-based morphometry (TBM) by using T1 and T2-weighted scans simultaneously for the non-linear registration of scans to better account for the normal contrast changes that occur in this age range, as well as using age-appropriate templates as the target for this registration. This technique was applied to the study of autism spectrum disorder (ASD), with the results showing distinct differences in the growth trajectories of children with ASD when compared to controls in several regions across the brain. The second proposed technique focuses on later childhood and adolescence (4-23 years of age) and consists of a morphometric normalization technique that uses a large longitudinal normative database to establish normal growth trajectories for the volume of several regions in the brain, as well as at the voxel level. This technique was applied to study 16p11.2 copy number variants, with the results showing significant differences for the duplication and deletion cohorts already present at 4.5 years of age and remaining stable throughout childhood and adolescence. Finally, we propose an extension of this normalization technique that includes the addition of a second normative dataset and improved preprocessing, enabling normalization in studies with data acquired at both 1.5 and 3 T. This technique was applied to study pediatric onset multiple sclerosis (MS), with the results showing a lack of age expected growth in the MS patients in overall brain volume, and specifically in the thalamus, putamen, and globus pallidus"--