| Surgery is a very complicated skill for a physician to learn. Hands on training is indispensable to the proper acquisition of this skill. Allowing new students to practice on real patients introduces a certain amount of risk for complications. Furthermore, patient flow through the ER is unpredictable and, for those patients with more rare medical cases, infrequent. In an attempt to address these issues, many surgical simulators have been developed and in some cases successfully undergone validation studies. Surgical simulators facilitate countless practice opportunities on even the most unique and challenging procedures without risk to patients. In order for surgical simulators to be effective, they must facilitate the learning of skills that are both relevant and transferable to real world procedures. Simulators do not need to be totally immersive, but must be sufficiently immersive for successful training.;At Stanford, the Surgical Simulation Group has developed simulators for otologic and rhinologic surgery. These simulators consist of a simulation engine running in software, a visual user interface, and a haptic user interface. In many surgical procedures, the tactile cues that come from physically interacting with tissue and bone are critical to successful outcomes. Consequently, the role of haptic devices is significant in surgical simulators designed to train surgical motor skills.;This work addresses three aspects of haptic hardware. First, a new technique for evaluating haptic hardware is developed and validated. The technique is simple to perform and only requires very little instrumentation (an accelerometer) to evaluate any arbitrary haptic device. Furthermore, the technique is able directly guide hardware designs while showing a clear relationship to a variety of haptic applications.;Second, a novel haptic device for microsurgery was designed, built, and tested. It has 6 degrees of freedom and is redundant in both actuation and sensing. It uses a novel cable transmission and motor selection technique to meet the design requirements for this application. The resultant performance is an improvement in both backdriveability and scale-independent measures such as dynamic range and acceleration.;Lastly, guidelines for the placement of haptic hardware within a simulator are developed. In order to develop the guidelines, investigation into the coordinate system used by humans for both programming and transferring motor skills is conducted. Results show, in contrast to contemporary assumptions, that fine motions of the hand use a coordinate system that is cue-dependent, with a strong preference for intrinsic coordinates. |
