A New Landmark-Independent Tool for Quantifying and Characterizing Morphologic Variation.

摘要

As advances in multi-dimensional imaging are made, there is a growing need for tools that enable quantitative assessment of both embryonic and postnatal structure and form. To this end, this dissertation develops a landmark-independent, deformable-registration based framework for quantifying and characterizing morphologic variation. This framework can utilize 3D surface images generated by any multi-dimensional biomedical imaging modality and provides quantitative assessment of potentially biologically relevant changes in 3D shape whether natural, caused by mutation, or influenced by non-genetic factors. In this work, using different example datasets, the utility and sensitivity of the tools developed in this work are demonstrated: 1) to detect subtle morphological phenotypes in specimens; 2) to quantify morphologic change over a normal developmental time series of embryonic development; and 3) to determine the impact of transient localized genetic manipulation and environmental change on morphogenesis. The development of these tools will assist in defining trajectories of normal growth and help underpin hypotheses as to the developmental basis and the specific role of the causative gene in conditions causing disruptions to normal growth.