Charles Explorer logo
🇬🇧

Statistical Mesh Analysis of Longitudinal Shape Changes

Publication at Faculty of Science, First Faculty of Medicine, Faculty of Mathematics and Physics |
2013

Abstract

Three dimensional surface scanning and object representation by triangular meshes has become a standard procedure in many fields dealing with analysis of shape and its changes (medicine, anthropology). However, collecting such data for longitudinal studies (repeated measurements of specimen changing in time) is time consuming and expensive.

Finding correspondences between models of individual specimens and between stages of development of particular specimen is a fundamental step for their visual and statistical analysis. We implemented a method for finding mutually corresponding points in triangular meshes.

We used these correspondences for analysis of variability of changes between two stages of shapes represented this way as well as for confirmation of statistical significance and visualization of differences between these changes in various populations.