Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.790133
Title: Automated morphometry for mouse brain MRI through structural parcellation and thickness estimation
Author: Ma, D.
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
Quantitative morphometric analysis is an important tool in neuroimaging for the study of understanding the physiology of development, normal aging, disease pathology and treatment effect. However, compared to clinical study, image analysis methods specific to preclinical neuroimaging are still lacking. The aim of this PhD thesis is to achieve automatic quantitative structural analysis of mouse brain MRI. This thesis focuses on two quantitative methods which have been widely accepted as quantitative imaging biomarkers: brain structure segmentation and cortical thickness estimation. Firstly, a multi-atlas based structural parcellation framework has been constructed, which incorporates preprocessing steps such as intensity non-uniformity correction and multi-atlas based brain extraction, followed by non-rigid registration and local weighted multi-atlas label fusion. Validation of the framework demonstrated improved performance compared to single-atlas-based structural parcellation, as well as to global weighted multi-atlas label fusion methods. The framework has been further applied to in vivo and ex vivo data acquired from the same cohort so that the respective volumetric analysis can be compared. The results reveal a non-uniform distribution of volume changes from the in vivo to the post-mortem brain. In addition, volumetric analysis based on the segmented structures showed similar statistical power on in vivo or ex vivo data within the same cohort. Secondly, a framework to segment the mouse cerebellar cortex sublayers from brain MRI data and estimate the thickness of the corresponding layers has been developed. Application of the framework on the experimental data demonstrated its ability to distinguish sublayer thickness variation between transgenic strains and their wild-type littermate, which cannot be detected using full cortical thickness measurements alone. In conclusion, two quantitative morphometric analysis frameworks have been pre-sented in this thesis. This demonstrated the successful application of translational quantitative methods to preclinical mouse brain MRI.
Supervisor: Sebastien, O. ; Lythgoe, M. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.790133  DOI: Not available
Share: