Visualization methods for analysis of 3D multi-scale medical data
Ricardo Millán, Leibniz Universität Hannover
This thesis was developed within the framework of the EU MultiScaleHuman Project, whose central aim lies on improving the current understanding of physiological human articulation with the application of supporting the diagnosis and treatment of musculoskeletal diseases. Towards this goal, this work contributes several methods in order to improve the visualization and processing of biomedical data. It especially addresses scientists whose workflow involves potentially massive amount of biomedical information.
In order to overcome the limitations in the current state of the art, this work contributes concepts for multi-scale visualization, particularly providing a 3D multilayered environment, in which global and detailed information from multi-scale biomedical data can be presented and explored, a combination of scientific and information visualization, which aims to overcome the limitation of using one group of visualization methods to represent biomedical information with different levels of abstraction, the introduction of semantic information from the multi-scale data for the creation, customization and enrichment of the multi-scale visualization, and a set of enriched intra-scale visualizations, enhancing the presentation and analysis of data within the considered scale in order to support the clinical workflow. In this context, this thesis presents concepts for the analysis of kinematical data of human joint articulation at the behavioral scale.
The multi-scale visualization framework has the potential to amplify the understanding of biomedical processes in the human body. This has been confirmed in an application scenario focusing on the investigation of knee joint kinematics based on experimental data collected in collaboration with MHH-LBB, the latter being a project partner within the MultiScaleHuman project. It was demonstrated, that the developed processing methods relying on Lie group theory and the corresponding visualization techniques allow for an intuitive interpretation of helical axes and their geometrical relation with respect to the knee joint anatomy, an efficient differentiation among several cases and states of knee joint, and a flexible exploration of the discussed data sets and of related data, thereby supporting the clinical workflow during the analysis of the knee joint articulation.