Development of a Reality-based, Haptics-enabled Simulator for Tool-tissue Interactions

Title of dissertation: DEVELOPMENT OF A REALITY-BASED, HAPTICS-ENABLED SIMULATOR FOR TOOL-TISSUE INTERACTIONS Kevin Lister, Doctor of Philosophy, 2012 Dissertation directed by: Professor Jaydev P. Desai Department of Mechanical Engineering The advent of complex surgical procedures has driven the need for finite element based surgical training simulators which provide realistic visual and haptic feedback throughout the surgical task. The foundation of a simulator stems from the use of accurate, reality-based models for the global tissue response as well as the tool-tissue interactions. To that end, ex vivo and in vivo tests were conducted for soft-tissue probing and in vivo tests were conducted for soft-tissue cutting for the purpose of model development. In formulating a surgical training system, there is a desire to replicate the surgical task as accurately as possible for haptic and visual realism. However, for many biological tissues, there is a discrepancy between the mechanical characteristics of ex vivo and in vivo tissue. The efficacy of utilizing an ex vivo model for simulation of in vivo probing tasks on porcine liver was evaluated by comparing the simulated probing task to an identical in vivo probing experiment. The models were then further improved upon to better replicate the in vivo response. During the study of cutting modeling, in vivo cutting experiments were performed on porcine liver to derive the force-displacement response of the tissue to a scalpel blade. Using this information, a fracture mechanics based approach was applied to develop a fully defined cohesive zone model governing the separation properties of the liver directly in front of the scalpel blade. Further, a method of scaling the cohesive zone parameters was presented to minimize the computational expense in an effort to apply the cohesive based cutting approach to real-time simulators. The development of the models for the global tissue response and local tooltissue interactions for probing and cutting of soft-tissue provided the framework for real-time simulation of basic surgical skills training. Initially, a pre-processing approach was used for the development of reality-based, haptics enabled simulators for probing and cutting of soft tissue. Then a real-time finite element based simulator was developed to simulate the probing task without the need to know the tool path prior to simulation. DEVELOPMENT OF A REALITY-BASED, HAPTICS-ENABLED SIMULATOR FOR TOOL-TISSUE INTERACTIONS