Towards Cognition-Guided Patient-Specific FEM-Based Cardiac Surgery Simulation

Biomechanical surgery simulation can provide surgeons with useful ancillary information for intervention planning, diagnosis and therapy. The simulation therefore most importantly needs to be patient-specific, surgical knowledge-based and comprehensive in terms of the underlying simulation model and the patient's data. Moreover, the simulation setup and execution should be largely automated and integrated into the surgical treatment workflow. However, this still rarely holds and simulation-based surgery support is not yet commonly established in the clinic. In this work, we address this problem in the context of cardiac surgery, and present the setup and results of a prototypic cognition-guided, patient-specific FEM-based cardiac surgery simulation system. We have designed a semantic data infrastructure and implemented cognitive software components that autonomously interact with the medical data via a common ontology. Using this setup, we anable the creation of knowledge-based, patient-specific surgery simulation scenarios for mitral valve reconstruction surgery, that are executed by means of the FEM simulation software HiFlow(3). The obtained simulation results are provided to the surgeon in order to support surgical decision making.