Sastin, Dmitri
2024.
Investigation into short association fibre tractography methods using surface-based approach.
PhD Thesis,
Cardiff University.
Item availability restricted. |
PDF
- Accepted Post-Print Version
Restricted to Repository staff only until 19 September 2025 due to copyright restrictions. Download (123MB) |
|
PDF (Cardiff University Electronic Publication Form)
- Supplemental Material
Restricted to Repository staff only Download (146kB) |
Abstract
Short association fibres (SAF) are a large yet critically understudied part of the human connectome. Their proximity to the cerebral cortex poses unique challenges for neuroimaging research. These include mixed information from grey and white matter, overlapping fibres with diverse orientations from functionally distinct groups, and variable cortical folding between individuals. Most neuroimaging literature investigating white matter to date has focused on long range connections where these challenges are less pronounced. There is therefore an unmet need to understand the impact of these challenges on the superficial white matter and develop strategies better suited for investigating short association fibres specifically. This thesis addresses the listed issues by placing the white-grey boundary (WGB) at the core of SAF tractography. It introduces a WGB surface-based approach for seeding and filtering of SAF, comparing its benefits to traditional voxel-based methods. SAF data representation on the WGB is used for between-subject analyses. Differential domain analysis of seed distribution in the vicinity of the WGB is also developed and used to inspect the impact of different seeding parameters on tractograms. The advantages of surface-based methods are then integrated into anatomically constrained tractography (ACT) framework. Additionally, the impact of data acquisition parameters and upsampling on SAF tractograms is investigated. Results show that the surface-based approach produces longer SAF streamlines and broader cortical coverage while promoting connections between gyri, in keeping with histological data. These qualities are most evident in high spatial and angular resolution upsampled data. Reduced accuracy of seed placement on the WGB appears to increase projection-to-SAF streamline ratio while resulting in more gyral bias. The modified ACT framework successfully retains surface-based advantages in whole-brain tractograms. However, caution is warranted, given moderate reliability and notable variability in streamline density among individuals. The scarcity of the existing validation data is also emphasised.
Item Type: | Thesis (PhD) |
---|---|
Date Type: | Completion |
Status: | Unpublished |
Schools: | Medicine |
Date of First Compliant Deposit: | 24 September 2024 |
Last Modified: | 24 Sep 2024 08:57 |
URI: | https://orca.cardiff.ac.uk/id/eprint/172250 |
Actions (repository staff only)
Edit Item |