Filipe M M Marreiros1,2,3, Y Heuzé4, M Verius5, C Unterhofer6, W Freysinger7, W Recheis5. 1. Center for Medical Image Science and Visualization (CMIV), Linköping University, 581 85, Linköping, Sweden. filipe.marreiros@liu.se. 2. Department of Medical and Health Sciences (IMH), Linköping University, 581 85, Linköping, Sweden. filipe.marreiros@liu.se. 3. Department of Science and Technology (ITN) - Media and Information Technology (MIT), Linköping University, Linköping, Sweden. filipe.marreiros@liu.se. 4. University of Bordeaux, UMR 5199 PACEA, Bordeaux Archaeological Sciences Cluster of Excellence, Université de Bordeaux, Allée Geoffroy de Saint Hilaire Bat. B8, CS 50023, 33615, Pessac Cedex, France. 5. Department of Radiology, Innsbruck Medical University, Innrain 52, Christoph-Probst-Platz, 6020, Innsbruck, Austria. 6. Clinical Department of Neurosurgery, Innsbruck Medical University, Innrain 52, Christoph-Probst-Platz, 6020, Innsbruck, Austria. 7. Department of Otorhinolaryngology (ENT), Hearing, Speech and Voice Disorders, Innsbruck Medical University, Innrain 52, Christoph-Probst-Platz, 6020, Innsbruck, Austria.
Abstract
PURPOSE: The aim of this work was to introduce a computer-aided design (CAD) tool that enables the design of large skull defect (>100 [Formula: see text]) implants. Functional and aesthetically correct custom implants are extremely important for patients with large cranial defects. For these cases, preoperative fabrication of implants is recommended to avoid problems of donor site morbidity, sufficiency of donor material and quality. Finally, crafting the correct shape is a non-trivial task increasingly complicated by defect size. METHODS: We present a CAD tool to design such implants for the neurocranium. A combination of geometric morphometrics and radial basis functions, namely thin-plate splines, allows semiautomatic implant generation. The method uses symmetry and the best fitting shape to estimate missing data directly within the radiologic volume data. In addition, this approach delivers correct implant fitting via a boundary fitting approach. RESULTS: This method generates a smooth implant surface, free of sharp edges that follows the main contours of the boundary, enabling accurate implant placement in the defect site intraoperatively. The present approach is evaluated and compared to existing methods. A mean error of 89.29 % (72.64-100 %) missing landmarks with an error less or equal to 1 mm was obtained. CONCLUSION: In conclusion, the results show that our CAD tool can generate patient-specific implants with high accuracy.
PURPOSE: The aim of this work was to introduce a computer-aided design (CAD) tool that enables the design of large skull defect (>100 [Formula: see text]) implants. Functional and aesthetically correct custom implants are extremely important for patients with large cranial defects. For these cases, preoperative fabrication of implants is recommended to avoid problems of donor site morbidity, sufficiency of donor material and quality. Finally, crafting the correct shape is a non-trivial task increasingly complicated by defect size. METHODS: We present a CAD tool to design such implants for the neurocranium. A combination of geometric morphometrics and radial basis functions, namely thin-plate splines, allows semiautomatic implant generation. The method uses symmetry and the best fitting shape to estimate missing data directly within the radiologic volume data. In addition, this approach delivers correct implant fitting via a boundary fitting approach. RESULTS: This method generates a smooth implant surface, free of sharp edges that follows the main contours of the boundary, enabling accurate implant placement in the defect site intraoperatively. The present approach is evaluated and compared to existing methods. A mean error of 89.29 % (72.64-100 %) missing landmarks with an error less or equal to 1 mm was obtained. CONCLUSION: In conclusion, the results show that our CAD tool can generate patient-specific implants with high accuracy.
Entities:
Keywords:
Cranial reconstruction; Geometric morphometrics; Radial basis functions and thin-plate spline; Reconstructive surgery
Authors: Marc Anton Fuessinger; Steffen Schwarz; Carl-Peter Cornelius; Marc Christian Metzger; Edward Ellis; Florian Probst; Wiebke Semper-Hogg; Mathieu Gass; Stefan Schlager Journal: Int J Comput Assist Radiol Surg Date: 2017-10-28 Impact factor: 2.924