Semi-automatic 3D virtual surgery environment generation from operative surgery manuals

Recently, virtual surgery technologies have much progress and many simulators have been developed for education, planning rehearsal and so on. However, development of VR-based surgical simulator takes much more labor and cost not only for implementing simulation modules but also for setting surgical environment. Considering that operative surgery manuals describe the knowledge of manipulations and the needed sources in surgical procedure and could be of great help to facilitate the work of building 3D virtual surgery environment effectively. In this study, we use the natural language processing technology to extract the information contained in the operative surgery manual, such as, the anatomical objects and surgical procedures. Then, we propose a framework that combines the natural language processing techniques and the surgical simulation technology together. In the framework, combined with medical ontological technologies, NLP modules are adapted to extract surgical information and XML file were made, which then be utilized by simulation modules using a simulation library in different system configurations to generate the final surgery simulator.