Surface Glyphs for Visualizing Multimodal Volume Data

In this paper we present concepts for integrating glyphs into volumetric data sets. These concepts have been developed with the goal to make glyphbased visualization of multimodal volumetric data sets more flexible and intuitive. We propose a surface-based glyph placement strategy reducing visual clutter as well as image-space glyph aggregation. Thus the user is not distracted by unwanted clustering, and his focus of attention can rather be guided using appropriate visual appearances. Furthermore, we present techniques to make the setup of glyph-based visualizations more intuitive. These concepts have been integrated into a user interface which supports easy configuration and comparison of different glyph setups. Based on the chosen setup a visual legend is generated automatically to make a step towards quantitative visual analysis. We will discuss the placement strategy as well as the glyph setup process, explain the used rendering techniques and provide application examples of multimodal visualizations using the proposed concepts. Multimodal volume visualization has to deal with the proper integration of data obtained from different sources. In the medical domain acquisition of different modalities is about to become a daily routine since modern scanners such as PET/CT, PET/MRT or SPECT/CT can be used to capture multiple registered volume data sets. PET data sets which usually have a lower resolution than CT data sets represent a functional image, e.g., metabolism activity, while CT data sets provide a detailed morphological image. To benefit from both modalities, they have to be visualized simultaneously in an integrated manner. In contrast to these modalities additional data can be derived from a given volume data set and visualized with our technique. For instance, cardiac wall thickness or wall motion can be calculated from time-varying medical data sets. The use of multimodal volumetric data sets is manifold not only in medicine but also in areas like meteorology and seismology. Furthermore, many physical simulations, e.g., fluid dynamics or quality insurance simulations, produce multimodal volumetric data sets.