Fakultät Informatik

Advanced Volume Rendering Techniques for Surgical Training Environments

 J. Georgii, Rüdiger Westermann
Computer Graphics and Visualization Group, Technische Universität München, Germany

M. Eder, L. Kovacs, A. Schneider, M. Dobritz
Klinikum rechts der Isar, Technische Universität München, Germany

Background

Surgical training environments as well as pre- and intra-operative planning environments require physics-based simulation systems to achieve the required accuracy. However, physics-based simulation at interactive rates is still a challenge due to the complex equations involved in the solution process. On the other hand, these environments often require a fast and intuitive visualization of internal material properties such as tissue stress to allow the surgeon to track the tissue stress induced by an intervention. If tissue is stressed too much, this often implies additional risks for the patient during a real surgical intervention. By exploiting a multigrid approach for soft-tissue simulation based on finite elements in combination with novel GPU-based volume rendering techniques for unstructured finite- element grids, a simulation support system that addresses the needs in virtual surgery simulators has been developed. Interactive direct volume rendering enables the surgeon to obtain immediate feedback on tissue stress induced by the applied intervention even for large data-sets.

Images

+

On the left, the boundary of an undeformed liver is shown. On the right, the deformed liver is visualized by means of direct volume rendering. The internal stress induced by the deformation is highlighted with green color.

+

Left: The outer surface of the volumetric model used for the physics-based simulation of gravity is shown. Right: Visualization of internal stress values. Red indicates high tissue stress while green marks undeformed regions. Such a visualization can assist in plastic and reconstructive surgery planning.

We have presented a virtual surgery simulator for medical training and pre- and intra- operative planning. By exploiting a multigrid approach for soft-tissue simulation in combination with novel GPU-based volume rendering techniques for unstructured finite- element grids, a simulation support system that addresses the needs in virtual surgery simulators has been developed. The system can handle large and thus very detailed models composed of different material structures. Interactive direct volume rendering enables the surgeon to obtain immediate feedback on tissue stress induced by the applied intervention.

Associated publications

 Advanced Volume Rendering for Surgical Training Environments
J. Georgii, M. Eder, L. Kovacs, A. Schneider, M. Dobritz, R. Westermann, International Journal of Computer Assisted Radiology and Surgery (Proceedings of CARS) 2007 [Bibtex]

 

News

Matthias Niessner, our new Professor from Stanford University, offers a number of interesting topics for  master theses.

 

PhD positions on   Computational Fabrication and 3D Printing and  Photorealistic Rendering for Deep Learning and Online Reconstruction are available at the Computer Graphics & Visualization group.