Kazuhiko Shinohara, MD, PhD. School of Health Sciences, Tokyo University of Technology
Background and Objectives: Artificial intelligence (AI) has recently been receiving increasing attention in the medical field. Ontology is defined as explicit formal specification of terms or concepts in a domain and relations among them. The successful introduction of AI to the medical field requires ontological and semantic descriptions of clinical procedures. This study suggests a method of ontological analysis for endoscopic surgery and investigates the feasibility and challenges of applying an ontological approach to surgery.
Materials and Methods: We analyzed the surgeon’s technique during basic training in the box trainer, such as pegboard transfer, cutting pattern, and suturing, and described these as an ontological procedure. In addition, the network and connectivity between medical electronic (ME) devices, such as the electric surgical unit, endoscope unit, and anesthesia machine were ontologically investigated and described.
Results: Surgical maneuvers in the box trainer for endoscopic surgery were successfully classified and described with reference to the relevant ontological concepts. Ontological descriptions of the connection status of the patient, ME devices, and energy supply outlets were successfully designated.
Conclusion: Ontological descriptions of surgical procedures and the network of ME devices for endoscopic surgery are possible. Ontological descriptions of the procedures in endoscopic surgery can be applied in other areas, such as computerized evaluation of medical training and certification, navigation and automation systems for robotic surgery, and medical alert systems for patient safety. This study revealed the need for simple and practical methods of clinical ontology.
Presented at the SAGES 2017 Annual Meeting in Houston, TX.
Abstract ID: 92338
Program Number: P402
Presentation Session: Poster Session (Non CME)
Presentation Type: Poster