Rössling, IvoDornheim, JanaDornheim, LarsBoehm, AndreasPreim, BernhardK. Buehler and A. Vilanova2014-02-062014-02-0620111017-4656https://doi.org/10.2312/EG2011/med/001-004Visual exploration of CT and MRI datasets in clinical practice is still dominated by slice-based viewing. Volume rendering is now widely available but seen primarily as a tool for a fast overview, and only rarely as a visualization to directly support clinical decisions. Research projects aiming at advanced 3D visualizations, such as smart visibility and illustrative renderings, usually fail to meet clinical demands, since the visualizations are not dedicated to specific diagnostic or treatment planning questions. Moreover, they are unfamiliar to users who need reliable and familiar visualizations as a basis for their crucial decisions. Discussions with clinical practicians reveal that parameterization of visual effects is too cumbersome and resulting visualizations are often too complex. We describe and discuss long-term experiences on developing, testing, and refining image analysis and visualization techniques for ENT surgery planning based on CT data. While visual quality and a faithful rendition of spatial relations indeed are essential, it turned out to be superior to generate sequences of rather simple 3D visualizations directly supporting specific treatment questions instead of presenting many anatomic structures simultaneously. We report on the actual clinical use of the system and discuss how it changed the surgical planning workflow.