An observation support system with an adaptive ontology-driven user interface for the modeling of complex behaviors during surgical interventions

The field of surgical interventions emphasizes knowledge and experience; explicit and detailed models of surgical processes are hard to obtain by observation or measurement. However, in medical engineering and related developments, such models are highly valuable. Surgical process modeling deals with the generation of complex process descriptions by observation. This places high demands on the observers, who have to use a sizable terminology to denominate surgical actions, instruments, and patient anatomies, and to describe processes unambiguously. Here, we present a novel method, employing an ontology-based user interface that adapts to the actual situation and describe the principles of the system. A validation study showed that this method enables observers with little recording experience to reach a recording accuracy of >90%. Furthermore, this method can be used for live and video observation. We conclude that the method of ontology-supported recording for complex behaviors can be advantageously employed when surgical processes are modeled.