Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.606725
Title: Numerical modelling of cerebrospinal fluid flow in the human ventricular system based on 4-dimensional radial basis function interpolation of MRI data
Author: Thewlis, Jonathan
Awarding Body: University of Nottingham
Current Institution: University of Nottingham
Date of Award: 2013
Availability of Full Text:
Access through EThOS:
Abstract:
The function and flow behaviour of cerebrospinal fluid (CSF) within the central nervous system (CNS) has been the source of much discussion within the medical community. Understanding of this area has important implications for the treatment of several pathological conditions. A central area of unresolved debate concerns the driving forces for CSF motion. The relative magnitude and role of pulsation of different regions of the CNS has come under question. Modern magnetic resonance imaging (MRI) methods allow transient measurement of CSF flow velocities. This data may be used in the construction of numerical models of CSF flow. However, the small scale of flow requires high velocity resolution, resulting in reduced spatial and temporal resolution. This places limitations on the accurate formation of boundary conditions. No authors have accurately modelled pulsation of the ventricular walls. There exists therefore a need for a 4-dimensional interpolation technique to produce a continuous temporal and spatial velocity profile of CSF flow. This thesis describes a novel 4-dimensionallocal radial basis function (RBF) software tool for the interpolation of uniformly spaced, small scale MRI data. The .application of the tool is demonstrated firstly on MRI velocity data of CSF flow within the brain. The method is then extended to the interpolation of such data to the inlet boundary cell centres of a numerical model ofCSF flow, constructed from anatomical MRI data. The model is validated through comparison with the MRI results: An algorithm for the removal of rigid body motion effects due to movement of the test subject during scanning is also described. CSF flow variation in the interpolated MRI results indicates a pulsatile flow pattern with expansion and contraction of the lateral ventricles. The results of the numerical simulation show some discrepancy with the MRI data. However, peak flow velocities in the cerebral aqueduct are comparable with the literature. Due to the complex geometrical features of the ventricle system and the limitations of the interpolated MRI data, the numerical simulation has been found to be highly sensitive to the positioning of the inlet boundaries. Rigid body motion effects are found to be negligible for larger scale CSF flow but may affect ventricular wall pulsation. l
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.606725  DOI: Not available
Share: