Beta 1


Title Quantification of Craniofacial Growth in Mice with Craniofacial Dysmorphology Caused by the Crouzon Mutation Fgfr2C342Y
Author Thorup, Signe Strann (Image Analysis and Computer Graphics, Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
Supervisor Larsen, Rasmus (Image Analysis and Computer Graphics, Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
Institution Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark
Thesis level Master's thesis
Year 2008
Abstract The Crouzon syndrome is characterised by growth disturbances caused by premature fusion of the cranial growth zones (craniosynostosis). A mouse model with mutation Fgfr2C342Y , equivalent to the most common Crouzon syndrome mutation, has a phenotype showing many parallels to the human counterpart. Quantifying growth in the Crouzon mouse model could test hypotheses of the relationship between craniosynostosis and dysmorphology (abnormal morphology) and thus lead to better understanding of the causes of Crouzon syndrome as well as providing knowledge relevant for surgery planning in humans in the future. The present thesis extends the work of Hildur Olafsdottir who worked on quantifying differences between normal and Crouzon mice in her Ph.D. thesis. Automatic non-rigid volumetric image registration was applied to micro CT scans of ten 4-week and twenty 6-week euthanized mice for growth modeling. Each age group consisted of 50% normal (wild-type) and 50% Crouzon mice. Four 3D mean shapes (atlases), one for each mouse-type and age group were created. Extracting a dense field of growth vectors for each mouse-type; growth models were created using linear interpolation. Visualisations in the form of 3D animations where made. Spatial regions of significantly different growth were identified using the local False Discovery Rate method (FDR), which estimates the expected percentage of false predictions in a set of predictions, and a Multivariate Analysis of Variance (MANOVA), which unlike local FDR incorporated the correlation between the x, y and z component of the growth vector. Additionally, after automatically locating anatomical landmarks; growth was estimated automatically for clinical interesting variables such as skull length, width and height. Furthermore, significant growth differences were found between Crouzon and normal mice, as well as significant differences in spatial distribution of growth. The work was extended to include an estimation of the change in asymmetry between 4 and 6-week. The reason for this was the hypothesis, that the Crouzon mice would develop a higher degree of asymmetry during growth. Thus, asymmetry analysis is a natural continuation of growth modeling.
Series IMM-M.Sc.-2008-108
Fulltext
Original PDF ep08_108_net.pdf (104.14 MB)
Admin Creation date: 2008-12-17    Update date: 2008-12-17    Source: dtu    ID: 231912    Original MXD