We present our experience in fast prototyping of a series of important but computation-intensive functionalities in surgical simulators based on newly released PhysX-enabled GPU. We focus on soft tissue deformation and bleeding simulation, as they are essential but have previously been difficult to be rapidly prototyped. A multilayered soft tissue deformation model is implemented by extending the hardware-accelerated mass-spring system (MSS) in PhysX engine. To ensure accuracy, we configure spring parameters in an analytic way and integrate a fast volume preservation method to overcome the volume loss problem in MSS. Fast bleeding simulation with consideration of both patient behavior and mechanical dynamics is introduced. By making use of the PhysX built-in SPH-based fluid solver with careful assignment of parameters, realistic yet efficient bleeding effects can be achieved. Experimental results demonstrate that our approaches can achieve both interactive frame rates and convincing visual effects even when complex models are involved.
|Name||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Conference||5th International Conference on E-learning and Games, Edutainment 2010|
|Period||1/08/10 → 1/08/10|
- Theoretical Computer Science
- Computer Science(all)