Mapping connections in the neonatal brain with magnetic resonance imaging

Thompson, Elinor (2021) Mapping connections in the neonatal brain with magnetic resonance imaging. PhD thesis, University of Nottingham.

PDF (Thesis - as examined) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Available under Licence Creative Commons Attribution.
Download (70MB) | Preview


The neonatal brain undergoes rapid development after birth, including the growth and maturation of the white matter fibre bundles that connect brain regions. Diffusion MRI (dMRI) is a unique tool for mapping these bundles in vivo, providing insight into factors that impact the development of white matter and how its maturation influences other developmental processes. However, most studies of neonatal white matter do not use specialised analysis tools, instead using tools that have been developed for the adult brain.

However, the neonatal brain is not simply a small adult brain, as differences in geometry and tissue decomposition cause considerable differences in dMRI contrast. In this thesis, methods are developed to map white matter connections during this early stage of neurodevelopment. First, two contrasting approaches are explored: ROI-constrained protocols for mapping individual tracts, and the generation of whole-brain connectomes that capture the developing brain's full connectivity profile. The impact of the gyral bias, a methodological confound of tractography, is quantified and compared with the equivalent measurements for adult data.

These connectomes form the basis for a novel, data-driven framework, in which they are decomposed into white matter bundles and their corresponding grey matter terminations. Independent component analysis and non-negative matrix factorisation are compared for the decomposition, and are evaluated against in-silico simulations. Data-driven components of dMRI tractography data are compared with manual tractography, and networks obtained from resting-state functional MRI. The framework is further developed to provide corresponding components between groups and individuals. The data-driven components are used to generate cortical parcellations, which are stable across subjects.

Finally, some future applications are outlined that extend the use of these methods beyond the context of neonatal imaging, in order to bridge the gap between functional and structural analysis paradigms, and to chart the development of white matter throughout the lifespan and across species.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Sotiropoulos, Stamatios N.
Bastiani, Matteo
Keywords: Infant, newborn; Neurodevelopment; Neonatal imaging; Diffusion MRI; White matter; Brain connectivity
Subjects: W Medicine and related subjects (NLM Classification) > WS Pediatrics
Faculties/Schools: UK Campuses > Faculty of Medicine and Health Sciences > School of Medicine
Item ID: 64848
Depositing User: Thompson, Elinor
Date Deposited: 04 Aug 2021 04:41
Last Modified: 04 Aug 2021 04:41

Actions (Archive Staff Only)

Edit View Edit View