Model-Based Visualization for Intervention Planning

Abstract

Computer support for intervention planning is often a two-stage process: In a first stage, the relevant segmentation target structures are identified and delineated. In a second stage, image analysis results are employed for the actual planning process. In the first stage, model-based segmentation techniques are often used to reduce the interaction effort and increase the reproducibility. There is a similar argument to employ model-based techniques for the visualization as well. With increasingly more visualization options, users have many parameters to adjust in order to generate expressive visualizations. Surface models may be smoothed with a variety of techniques and parameters. Surface visualization and illustrative rendering techniques are controlled by a large set of additional parameters. Although interactive 3d visualizations should be flexible and support individual planning tasks, appropriate selection of visualization techniques and presets for their parameters is needed. In this chapter, we discuss this kind of visualization support. We refer to model-based visualization to denote the selection and parameterization of visualization techniques based on 'a priori knowledge concerning visual perception, shapes of anatomical objects and intervention planning tasks.