Virtual Reality-based Medical Training Simulator for Bilateral Sagittal Split Osteotomy


This project aims at developing a specialized training system dedicated to the teaching of a major maxillofacial surgery procedure based on Virtual Reality technology. The medical procedure of interest, the bilateral sagittal split osteotomy (BSSO), allows the displacement of the lower jaw in all three spatial dimensions, e.g., to correct an under- or overbid. The procedure is performed via an intraoral approach and includes the creation of an osteotomy line in the mandible using a saw or a burr, followed by a controlled splitting of the bone by a reversed twisting of one or two chisels inserted into the line.

  Screenshot of a BSSO simulation Copyright: © VR Group

(Left) Screenshot of a BSSO simulation including head in operation pose and surgical instruments. (Right) Screenshot of a BSSO simulation including chisels inserted in the osteotomy line of the bone and a visualization of interaction forces (green and blue lines). Here, the bone already started to split; the crack is visible inside the bone in red.

To enable a realistic and efficient simulation of the breaking bone in real time, the extended finite element method is employed as it allows the effective modeling of deformations and induced structural changes. Furthermore, as one important aspect of the simulator is training specific motor abilities for the split osteotomy, one prerequisite to the system is a reproduction of the real interaction with a high degree of realism. Therefore, the simulator utilizes multi-modal human-machine interfaces, i.e., haptic devices allowing a trainee to control virtual surgical instruments intuitively and provide him with force feedback. To enable a stable and responsive haptic interaction, the scene would need to be simulated at a haptic update rate, which is usually considered to be around 1000 Hz. Nevertheless, some components of the simulation are very complex, i.e., simulating the bone with an adequate mesh resolution results in a much slower simulation rate. To resolve this conflict, a multirate simulation approach is applied where multiple simulation components work together but utilize different representations of the simulated scene objects, use models of their physical behavior with different complexity, and run at different update rates. As foundation for this physics simulation experimentally determined material parameters are used together with anatomical models extracted from cone beam CT-Scans.


Further Information

  1. Joint work with the RWTH Chair for Computational Analysis of Technical Systems and the Department of Oral and Maxillofacial Surgery (only in german), University Hospital Aachen
  2. Funded by the German Research Foundation