Abstract:
The subject of this thesis was to determine the clinical validity of a CT (computer tomography)-based simulation of the implantation of an electronic hearing device in a VR surrounding with an already introduced, reasonable and commercially available VR system. We employed the system for a series of 29 temporal bones of human cadavers and for a series of 25 patients.
After processing the informations based on CT-scans, the resulting three-dimensional surfaces were transferred to two different test procedures. At first the dimension of the anatomical correspondence between the real and the virtual mastoidectomy was determined by the "surfacer test". After that, in the "fitting-test", it was possible to simulate the implantation process in the VR surrounding by bringing together the CAD data of the hearing aid modules and the contours of the temporal bones after virtual mastoidectomy. Afterwards the results concerning the implantability in the VR environment and the implantability in the real operation were compared.
A high registration quality between the resulting surfaces of the temporal bones which were operated in the virtual reality surrounding and the temporal bones which were operated in reality was reached for all patients. This was the basis for the usability of the "surfacer test". The "surfacer test" itself proved the excellent correspondence between the real and the virtual mastoidectomy. In those for the implantation relevant regions of the temporal bone the anatomical differences between the real and the virtual mastoidectomy were predominantly maximally 1 mm.
The "fitting test" took on an average 90 minutes, including the time, which was needed to process and to transfer the CT based information into the VR environment. The simulation of the implantation itself could be visualized in real time. Regarding the series of temporal bones of human cadavers and the series of patients, the prediction of the "fitting test" concerning the implantability was in 94% correct. Difficulties with the installation or reasons for a failure of the operation could be predicted precisely.
The used CT based VR system has turned out in this work well servable and stable. The simulation results were very precise and showed a high prediction exactness. The excellent clinical applicability of the VR system in and anatomically complicated an interindividually very variable body region for the preoperative planning and simulation could be indicated.
Therefore, the advancement of the used VR system and its application for preoperative planning in other body regions seems recommendable. In future the system should be improved by summarizing single working steps in order to reduce the time which is needed to process the data. Besides, the aim should be the integration of three dimensional visualization technologies in order to simplify the exact localization of objects in VR surroundings.