Semedo, Carla; (2021) Exploring thalamic nuclei segmentation based on structural connectivity. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text Carla_Semedo_final_PhD_thesis.pdf Download (15MB) | Preview

Abstract

The thalamus is a deep grey matter structure that relays information between subcortical and cortical regions, especially sensory, motor and limbic signals. It is composed of distinct nuclei that have unique patterns of structural connectivity. The thalamic nuclei have distinct functions, so their accurate identification has multiple advantages. First, it can enable the inference of any anatomical and functional changes induced by a particular neurological disorder, promoting a better understand of the disease mechanism and development of biological markers. Additionally, it can improve surgical planning and electrodes placement in deep brain stimulation (DBS) interventions, minimising any adverse effects. This structure is easily identified in conventional imaging modalities such as Computed Tomography (CT) and structural Magnetic Resonance (MR), however there is no enough contrast to differentiate between thalamic nuclei. Diffusion weighted imaging (DWI) has emerged as an alternative modality, which assesses water molecules diffusion within biological tissues, providing useful information about tissue microstructure and fibre architecture. Thalamic nuclei have distinct myelo-architecture, which translate into unique diffusion properties. Thus, the majority of thalamus parcellation strategies reported until the present date rely on DW data to divide the thalamus into its distinct nuclei by considering local diffusion properties and/or structural connectivity. These approaches demonstrated the potential of \textit{in vivo} resolving thalamic nuclei, however advances can still be made. Future thalamus parcellation procedures may focus on more complex clustering techniques, structural priors and robust MR-based features. In this context, this project aimed the development of an automatic thalamus parcellation strategy to segment thalamic nuclei by considering DWI $-–$ local fibre orientation and structural connectivity (tractography and anatomical priors). In terms of its translation, it can be used to better detect thalamic nuclei and infer any morphometric changes due to neurological disorders, with consequent improvement in understanding a disease mechanism, its diagnosis and treatment.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item