Heterogeneous Deformable Object Modeling for Medical Surgical Simulation and Collaborative Product Development with Haptic Interfaces

Show full item record

Title: Heterogeneous Deformable Object Modeling for Medical Surgical Simulation and Collaborative Product Development with Haptic Interfaces
Author: Lin, Shiyong
Advisors: Dr. Ezat T. Sanii, Committee Member
Dr. Christopher G. Healey, Committee Member
Dr. Roger J. Narayan, Committee Co-Chair
Dr. Yuan-Shin Lee, Committee Chair
Abstract: This research focuses on the investigation of heterogeneous deformable object modeling, physically based simulation, haptic force rendering and collaborative techniques for medical surgical simulation, virtual prototyping and collaborative product development. Heterogeneous deformable models can be used to present internal geometric structures and different material properties of biological tissues and other soft-material objects for many Virtual Reality (VR) systems like surgical simulators. Cutting simulation is an important component of VR systems to modify the topology of deformable models. In this paper, a tri-ray node snapping algorithm is presented to generate volumetric heterogeneous mass spring models from a set of interface surfaces between different materials of deformable objects. A constrained local static integration method is proposed to quickly find the equilibrium solution of physically-based deformation behaviors. A 3D node snapping algorithm is developed to implement the topology modification on heterogeneous deformable models. Smooth cut is generated directly by duplicating and displacing mass points that are snapped along cutting planes. Sets of triangular surfaces representing different soft tissues are generated along the new cut to present internal geometric structures and corresponding material properties. A quasi-static algorithm is presented to refine the mesh in the vicinity of new cuts. A lab-built 6-DOF (degree of freedom) input and 5-DOF output haptic interface system is integrated with the developed system to provide force-torque feedback to users. The marriage of network collaborative technology and the proposed deformable object modeling and simulation techniques has a great potential for more extensive applications. Challenges such as heavy computation and data synchronization exist when integrating heterogeneous deformable object models and haptic interfaces with collaborative VR systems. To balance the computational burden of haptic rendering and deformable object simulation, a hybrid network architecture is proposed in this paper. An adaptive artificial time compensation method is developed for the collaborative haptic VR system to reduce the time discrepancy between the server and the clients. Interpolation and extrapolation approaches by Verlet integration are used to synchronize graphic and haptic data transmitted over the network. The results show that the presented heterogeneous deformable modeling, haptic rendering and collaborative techniques can be used in medical simulation, product design and manufacturing, virtual prototyping, computer games and collaborative VR applications with enhanced visual and tactile virtual environments.
Date: 2008-11-07
Degree: PhD
Discipline: Industrial Engineering
URI: http://www.lib.ncsu.edu/resolver/1840.16/4650

Files in this item

Files Size Format View
etd.pdf 2.456Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record