Towards semantically steered navigation in shape spaces exemplified by rodent skull morphology in correlation to external attributes
Final Report Abstract
In biology and medicine, understanding the variability of an anatomy sampled from a population is key to distinguish (population-)normal and pathological morphology and uncover correlation to extrinsic factors. In order to gain such insights, visualization plays a central role, especially if anatomy is assessed via medical imaging. This project contributes several novel methods for investigating anatomic variability in 3D images via interactive displays. In combination, these methods facilitate a visual analytics system that enables unconstrained exploration as well as targeted analysis of anatomic variability, co-variation between different structures on the same anatomy and correlation of its form to extrinsic attributes. Following D’Arcy Thompsons classical “method of transformations”, deformations are used to encode differences in form and finally describe variation with respect to a template anatomy, that represents the statistical first moment or mean. The work at hand features accurate modeling and interactive rendering of deformations at high spatial resolution. The very unique collection of high quality CT datasets of rodent skulls contributed by Dr. Schunke provided an ideal testbed for our methods and her continuous feedback was invaluable during development and evaluation of the novel visual analytics methods. Exemplary investigations on the influence of phylogeny, diet and geography on the form of rodent skull and mandible were conducted. In the future we hope to see applications of our methods in morphometrics and computational anatomy. Population studies provide another ideal application domain because of their exploratory nature. However, for the ensemble sizes treated there, improving scalability of our methods does require additional research.
Publications
- Semantically steered visual analysis of highly detailed morphometric shape spaces. In BioVis 2011: 1st IEEE Symposium on biological data visualization, pages 151–158, October 2011
Max Hermann, Anja C. Schunke, and Reinhard Klein
(See online at https://doi.org/10.1109/BioVis.2011.6094060) - Interactive steering of mesh animations. In ACM SIGGRAPH/Eurographics Symposium on Computer Animation, July 2012
Anna Vögele, Max Hermann, Björn Krüger, and Reinhard Klein
(See online at https://doi.org/10.2312/SCA/SCA12/053-058) - A visual analytics approach to study anatomic covariation. In IEEE PacificVis 2014, March 2014
Max Hermann, Anja C. Schunke, Thomas Schultz, and Reinhard Klein
(See online at https://doi.org/10.1109/PacificVis.2014.53) - Anatomic segmentation of statistical shape models. Symposium on Statistical Shape Models & Applications (SHAPE2014) in Delémont, Switzerland, June 2014
Max Hermann, Anja C. Schunke, and Reinhard Klein
- Efficient unsupervised temporal segmentation of human motion. In ACM SIGGRAPH/Eurographics Symposium on Computer Animation, July 2014
Anna Vögele, Björn Krüger, and Reinhard Klein
(See online at https://doi.org/10.2312/sca.20141135) - A visual analytics perspective on shape analysis: State of the art and future prospects. Computers & Graphics, 53, Part A:63–71, September 2015
Max Hermann and Reinhard Klein
(See online at https://doi.org/10.1016/j.cag.2015.08.008) - Accurate interactive visualization of large deformations and variability in biomedical image ensembles. IEEE Transactions on Visualization and Computer Graphics, 22(1):708–717, January 2016
Max Hermann, Anja C. Schunke, Thomas Schultz, and Reinhard Klein
(See online at https://doi.org/10.1109/TVCG.2015.2467198) - Visual analytics methods for shape analysis of biomedical images exemplified on rodent skull morphology. PhD thesis, University of Bonn, Germany, 2017
Max Hermann
