The generation of multimodal virtual environments for surgical training is complicated by the necessity to develop heterogeneous simulation scenarios such as surgical incision, cauterization, bleeding, and smoke generation involving the interaction of surgical tools with soft biological tissues in real time. While several techniques ranging from rapid but nonphysical geometry-based procedures to complex but computationally inefficient finite element analysis schemes have been proposed, none is uniquely suited to solve the digital surgery problem. In this paper we discuss the challenges facing the field of realistic surgery simulation and present a novel point-associated finite field (PAFF) approach, developed specifically to cope with these challenges. Based upon the equations of motion dictated by physics, this technique is independent of the state of matter, geometry and material properties and permits different levels of detail. We propose several specializations of this scheme for various operational complexities. The accuracy and efficiency of this technique is compared with solutions using traditional finite element methods and simulation results are reported on segmented models obtained from the Visible Human Project.
BackgroundSurgical teaching has been based traditionally on the preceptor or apprenticeship model, in which the novice surgeon learns with small groups of peers and superiors, over time, in the course of patient care. The operating room and the patient, however, comprise the most common, the most readily available, and often the only setting where hands-on training takes place. The novice surgeon acquires skills by first observing experienced surgeons in action and then by progressively performing, under varying degrees of supervision, more of the surgical procedures, as his/her training advances and his/her skill level increases.