Lap Nephrectomy Simulator

 

Background/Problem

Patient information specific bio-physical based real time surgery simulations for residents/students and surgeon's education and training practices before surgery not only have immense learning benefits but also have practical value.
This effort focuses attention on challenges and the path forward towards the development of bio-physical based real time surgery simulations with particular attention to laparoscopic nephrectomy.
To achieve realism with real time computations poses significant technical challenges. To include careful considerations as related to geometric modeling and mesh generation, numerical selection and treatment of the space discretization process, design of time integrators with adaptive time step features to enable real time computations of the dynamic equations of motion, accurate material modeling of human soft tissues and organs and the like, realism of surgical procedures associated with surgical instruments with tissues and organs and modeling of damage with contact detection, collision and response, and integration with graphic visualization.

Features of the Simulation Project:

• Renal anatomic learning modules
• Bio-physically based retroperitoneal model for the METI SurgiSim platform
• Real time performance
• Bio-physically based deformation
• Human soft-tissue model from biomechanical property experiments
• Accurate/realistic interaction

Challenges in General

• VR Surgery is one the most complicated virtual environment scenarios (soft-tissue)
• Demands realistic algorithms to deliver accurate interaction results in real time
• Precise physical model of organs are not readily available
• The simulation of the behavior of tissue has a high computation burden
• Material models and characterization

Approaches/Solutions

Material model:

• Perform experiments on human tissues to extract material properties

Space aspects:

• Hybrid approach: Develop Beam-Model in conjunction with mass-spring model, FEM model, Meshfree model if necessary, Large deformation and nonlinear aspects
• Reduced order models which retain the rich physics

Time aspects

• Efficient Forward displacement time integration methods for computational efficient and suitable for accurate modeling of dynamic interactions
• Collision detection and collision response
• Spatial tessellation based method as fast and robust collision detection to handle complicated VR scenarios

Team Members:

• Dr. Rob Sweet
• Prof. Kumar Tamma
• Prof. Desong Sha
• Dr. Xiangmin Zhou
• Dr. Yunhe Shen
• Dr. Ravi Chityala
• Shalom Lidgi
• Dr. Nan Zhang
• Dan Burke

Project View

Hilar Models

Hilar Mesh and Tool

Virtual Tissue Interaction