Computationally efficient techniques for real time surgical simulation with force feedback

In this paper, we present computationally efficient algorithms for the real time simulation of minimally invasive surgical (MIS) procedures. To develop a surgical simulator for training medical personnel, due to high computational speed required for realistic simulation in real time, there is a fundamental trade-off between realism and processing speed of the simulator. Our research focuses on how we can optimally utilize computing resources in the presence of constraints related to real time performance. We first present a novel approach, to rapid collision detection so as to maximize the available time for computing collision response. We then present a real time tissue model that computes visual rendering of deformations and haptic rendering of interaction forces by using a newly developed meshless numerical scheme. By integrating these techniques with a force-feedback device and a visual display connected to PC, we simulate a specific procedure (palpation) with update rate of 1 KHz for force and 30 Hz frame rate for graphics.