Construction of a Knowledge Base for Empirical Knowledge in Neurosurgery

Neurosurgeons accumulate a variety of empirical knowledge through surgeries. Post-operative reports, incident reports, and accident reports are effective means of recording and sharing empirical knowledge, but they are costly to analyze them, and new methods of sharing knowledge are needed. In addition, the interface using CG technology, which is used in surgical planning, is actively used as a means for doctors to easily obtain information, but it is not widely used for knowledge sharing. In this research, we aim to build a knowledge base to convey and utilize physicians’ know-how, which is difficult to convey, by accurately expressing empirical knowledge using CG technology. One of the challenges in sharing empirical knowledge is the difficulty of handling medical information from the viewpoint of personal information protection. Medical information is data that can easily identify individuals and that has a very high importance of rare data. Therefore, we examined the environment for using medical information and appropriate anonymization. First, we proposed a method for constructing an ontology for neurosurgery based on the medical ontology that has been studied in the medical field by organizing the structure of empirical knowledge of doctors. In addition, we designed and fabricated a prototype interface using 3D models as a system for data input and search display. We selected the glTF format as the 3D model format to be used in this study. In this paper, we report on the construction of a knowledge base for sharing empirical knowledge of neurosurgery, and the evaluation of the ontology constructed by the proposed method.