在神經外科中，經由顱骨切開術產生的切口可以由顱骨成形術(一種使用植入物修復顱骨缺損的外科手術)來修復。然而，因為難以完美符合顱骨切口，外科醫生便經常在移植預製的顱片(版)到缺損區域時遭遇困難。使用鏡像成形、表面內插、或是模板形變的先前研究在顱骨重建時符合病患原始外觀的需求上總是多所受限，而在本論文中，我們使用病患自己的低解析度供診斷用與高解析度顱骨缺損之電腦斷層影像，一併保有其原始外形。由於顱骨重建的正確性與低解析度影像上的部份體積效應以及高解析度影像上的顱骨缺損比例有關，擁有完整顱骨的低解析度影像先重新取樣、取閥值然後做動態輪廓模型來削減部份體積假影，產生的低解析度影像與擁有不同比例顱骨缺損的高解析度影像對位後就可取出顱骨切開部位。我們進一步發展使用三維多重網格法的可形變表面模型與利用多重解析度影像對位來增進計算效率。為了估測演算法效能，我們使用一組顱骨假體來模擬六種顱骨缺陷狀況，其中包含百分之二十、四十比例的單側、雙側與跨中線缺損。修補顱骨影像的全部處理時間為二十分鐘左右，而其整體重建可達到次體素正確性。實驗結果說明我們提出的演算法可以拿來有效地製造自訂的植入物，並且在臨床顱骨成形術上使用。

In neurosurgery, cranial incisions during craniotomy can be recovered by cranioplasty—a surgical operation using cranial implants to repair skull defects. However, surgeons often encounter difficulties when grafting prefabricated cranial plates into defective areas, since a perfect match to the cranial incision is difficult to achieve. Previous studies using mirroring technique, surface interpolation, or deformed template had limitations in skull reconstruction to match the patient’s original appearance. In this thesis, we utilized low-resolution diagnostic and high-resolution defective computed tomography images from the patient to repair skull defects, whilst preserving the original shape. Since the accuracy of skull reconstruction was associated with the partial volume effects in the low-resolution images and the percentage of the skull defect in the high-resolution images, the low-resolution images with intact skull were resampled and thresholded followed by active contour model to suppress partial volume artifacts. The resulting low-resolution images were registered with the high-resolution ones, which exhibited different percentages of cranial defect, to extract the incised cranial part. We further developed the deformable surface model using 3D multigrid techniques and employed the multiresolution image resgistration to improve the computational efficiency. To evaluate the performance of the proposed algorithm, a set of skull phantoms were manufactured to simulate six different conditions of cranial defects, namely, unilateral, bilateral, and cross-midline defects with 20% or 40% skull defects. The overall image processing time in reconstructing the repaired skull images required about 20 minutes, and the overall reconstruction reached subvoxel accuracy. Experimental results demonstrated that the proposed algorithm effectively created a customized implant, which can readily be used in clinical cranioplasty.

[1] Adams R, Bischof L (1994) Seeded region growing. IEEE Trans Pattern Anal Mach Intell 16:641–647.
[2] Agner C, Dujovny M, Evenhouse R, Charbel FT, Sadler L (1998) Stereolithography for posterior fossa cranioplasty. Skull Base Surg 8:81–86.
[3] Barrett R (1994) Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods. Society for Industrial and Applied Mathematics, Philadelphia .
[4] Besl PJ, Mckay ND (1992): A method for registration of 3-D shapes. IEEE Trans Pattern Anal Mach Intell 14:239–256.
[5] Borgefors G (1988): Hierarchical chamfer matching: a parametric edge matching algorithm. IEEE Trans Pattern Anal Mach Intell 10:849–865.
[6] Briggs WL, Henson VE, McCormick SF (2000) A Multigrid Tutorial. 2nd ed. Society for Industrial and Applied Mathematics, Philadelphia.
[7] Bronstein MM, Bronstein AM, Kimmel R, Yavneh I (2006) Multigrid multidimensional scaling. Numer Linear Algebr Appl 13:149–171.
[8] Carr JC, Fright WR, Beatson RK (1997) Surface interpolation with radial basis functions for medical imaging. IEEE Trans Med Imaging 16:96–107.
[9] Chong CS, Lee HP, Kumar AS (2006) Automatic hole repairing for cranioplasty using Bezier surface approximation. J Craniofac Surg 17:344–352.
[10] Cignoni P, Rocchini C, Scopigno R (1998) Metro: measuring error on simplified surfaces. Comput Graph Forum 17:167–174.
[11] Cremers D, Tischhäuser F, Weickert J, Schnörr C (2002) Diffusion snakes: introducing statistical shape knowledge into the Mumford-Shah functional. Int J Comput Vis 50:295–313.
[12] Dean D, Min K-J (2003) Deformable templates for preoperative computer-aided design and fabrication of large cranial implants. Int Congr Ser 1256:710–715.
[13] D'Urso PS, Effeney DJ, Earwaker WJ, Barker TM, Redmond MJ, Thompson RG, Tomlinson FH (2000) Custom cranioplasty using stereolithography and acrylic. Br J Plast Surg 53:200–204.
[14] Frohn-Schauf C, Henn S, Witsch K (2004) Nonlinear multigrid methods for total variation image denoising. Compu Vis Sci 7:199–206.
[15] Garland M, Heckbert PS (1997) Surface simplification using quadric error metrics. In: Owen GS, Whitted T Mones-Hattal B (eds) Proceedings of the 24th Annual Conference on Computer Graphics and Interactive Techniques, Los Angeles, ACM Press/Addison-Wesley Publishing Co., New York, pp 209–216.
[16] Gholipour A, Kehtarnavaz N, Briggs R, Devous M, Gopinath K (2007) Brain functional localization: A survey of image registration techniques. IEEE Trans Med Imaging 26:427–451.
[17] Gonzalez RC, Woods RE (2008) Digital Image Processing. Pearson/Prentice Hall, Upper Saddle River.
[18] Gopakumar S (2004) RP in medicine: a case study in cranial reconstructive surgery. Rapid Prototyp J 10:207–221.
[19] Haber E, Modersitzki J (2006) A multilevel method for image registration. SIAM J Sci Comput 27:1594–1607.
[20] Han X, Xu C, Prince JL (2007) Fast numerical scheme for gradient vector flow computation using a multigrid method. IET Image Process 1:48–55.
[21] Hartkens T, Hill DLG, Castellano-Smith AD, Hawkes DJ, Maurer CR, Martin AJ, Hall WA, Liu H, Truwit CL (2003) Measurement and analysis of brain deformation during neurosurgery. IEEE Trans Med Imaging 22:82–92.
[22] Hutchinson P, Timofeev I, Kirkpatrick P (2007) Surgery for brain edema. Neurosurg Focus 22:E14.
[23] Hutton BF, Braun M, Thurfjell L, Lau DYH (2002) Image registration: an essential tool for nuclear medicine. Eur J Nucl Med Mol Imaging 29:559–577.
[24] Kale SC, Lerch JP, Henkelman RM, Chen XJ (2008) Optimization of the SNR-resolution tradeoff for registration of magnetic resonance images. Hum Brain Mapp 29:1147–1158.
[25] Kass M, Witkin A, Terzopoulos D (1987) Snakes: Active contour models. Int J Comput Vis 1:321–331.
[26] Kucukyuruk B, Abuzayed B, Sanus G, Aydin S, Aydin S (2011) Cranioplasty: review of materials and techniques. J Neurosci Rural Pract 2:162–167.
[27] Lee M-Y, Chang C-C, Lin C-C, Lo L-J, Chen Y-R (2002) Custom implant design for patients with cranial defects. IEEE Eng Med Biol 21:38–44.
[28] Lee S-C, Wu C-T, Lee S-T, Chen P-J (2009) Cranioplasty using polymethyl methacrylate prostheses. J Clin Neurosci 16:56–63.
[29] Li X, Zhang PP, Brisman R, Kutcher G (2005) Use of simulated annealing for optimization of alignment parameters in limited MRI acquisition volumes of the brain. Med Phys 32:2363–2370.
[30] Liao Y-L, Sun Y-N, Lu C-F, Wu Y-T, Wu C-T, Lee S-T, Lee J-D (2010) Skull-based registration of intra-subject CT images: The effects of different resolutions and partial contents. In: Mahadevan V, Zhou J (eds) Proceeding of the 2nd International Conference on Bioinformatics and Biomedical Technology (ICBBT), Research Publishing Services, Singapore, pp. 269–272.
[31] Liao Y-L, Lu C-F, Sun Y-N, Wu C-T, Lee J-D, Lee S-T, Wu Y-T (2011) Three-dimensional reconstruction of cranial defect using active contour model and image registration. Med Biol Eng Comput 49:203–211.
[32] Lorensen WE, Cline HE (1987) Marching cubes: A high resolution 3D surface construction algorithm. ACM SIGGRAPH Comput Graph 21:163–169.
[33] Luebke DP (2001) A developer's survey of polygonal simplification algorithms. IEEE Comput Graph Appl 21:24–35.
[34] Maes F, Collignon A, Vandermeulen D, Marchal G, Suetens P (1997) Multimodality image registration by maximization of mutual information. IEEE Trans Med Imaging 16:187–198.
[35] Maes F, Vandermeulen D, Suetens P (1999) Comparative evaluation of multiresolution optimization strategies for multimodality image registration by maximization of mutual information. Med Image Anal 3:373–386.
[36] Manawadu D, Quateen A, Findlay JM (2008) Hemicraniectomy for massive middle cerebral artery infarction: A review. Can J Neurol Sci 35:544–550.
[37] Maravelakis E, David K, Antoniadis A, Manios A, Bilalis N, Papaharilaou Y (2008) Reverse engineering techniques for cranioplasty: A case study. J Med Eng Technol 32:115–121.
[38] Melax S (1998) A simple, fast, and effective polygon reduction algorithm. Game Dev 5:44–49.
[39] Movassaghi K, Ver Halen J, Ganchi P, Amin-Hanjani S, Mesa J, Yaremchuk MJ (2006) Cranioplasty with subcutaneously preserved autologous bone grafts. Plast Reconstr Surg 117:202–206.
[40] Ohtake Y, Belyaev A, Bogaevski I (2001) Mesh regularization and adaptive smoothing. Comput Aided Design 33:789–800.
[41] Papandreou G, Maragos P (2007) Multigrid geometric active contour models. IEEE Trans Image Process 16:229–240.
[42] Pelizzari CA, Chen GTY, Spelbring DR, Weichselbaum RR, Chen CT: Accurate three-dimensional registration of CT, PET, and/or MR images of the brain. J Comput Assist Tomogr 13:20–26, 1989.
[43] Pham DL, Prince JL (1999) Adaptive fuzzy segmentation of magnetic resonance images. IEEE Trans Med Imaging 18:737–752.
[44] Pluim JPW, Maintz JBA, Viergever MA (2001) Mutual information matching in multiresolution contexts. Image Vis Comput 19:45–52.
[45] Press WH (1992) Numerical Recipes in C: The Art of Scientific Computing. Cambridge University Press, Cambridge/New York.
[46] Roche A, Malandain G, Pennec X, Ayache N (1998): The correlation ratio as a new similarity measure for multimodal image registration. In: Wells WM., Colchester A and Delp S (eds) First International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), LNCS 1496:1115–1124.
[47] Rusinek H, Tsui WH, Levy AV, Noz ME, Deleon MJ (1993) Principal axes and surface fitting methods for 3-dimensional image registration. J Nucl Med 34:2019–2024.
[48] Schirmer CM, Ackil AA, Malek AM (2008) Decompressive craniectomy. Neurocrit Care 8:456–470.
[49] Shi L, Yu Y, Bell N, Feng W-W (2006) A fast multigrid algorithm for mesh deformation. ACM Trans Graph 25:1108–1117.
[50] Sommer HJ, Eckhardt RB, Shiang TY (2006) Superquadric modeling of cranial and cerebral shape and asymmetry. Am J Phys Anthropol 129:189–195.
[51] Subramaniam S, Hill MD (2009) Decompressive hemicraniectomy for malignant middle cerebral artery infarction: An update. Neurologist 15:178–184.
[52] Sykes JR, Brettle DS, Magee DR, Thwaites DI (2009) Investigation of uncertainties in image registration of cone beam CT to CT on an image-guided radiotherapy system. Phys Med Biol 54:7263–7283.
[53] Taub PJ, Rudkin GH, Clearihue WJ, Miller TA (2003) Prefabricated alloplastic implants for cranial defects. Plast Reconstr Surg 111:1233–1240.
[54] Terzopoulos D (1983) Multilevel computational processes for visual surface reconstruction. Comput Vis Graph Image Process 24:52–96.
[55] Terzopoulos D (1986) Image analysis using multigrid relaxation methods. IEEE Trans Pattern Anal Mach Intell 8:129–139.
[56] Turkington TG, Hoffman JM, Jaszczak RJ, Macfall JR, Harris CC, Kilts CD, Pelizzari CA, Coleman RE (1995) Accuracy of surface fit registration for PET and MR brain images using full and incomplete brain surfaces. J Comput Assist Tomogr 19:117–124.
[57] van Herk M, Gilhuijs KGA, de Munck JC, Touw A (1997) Effect of image artifacts, organ motion, and poor segmentation on the reliability and accuracy of three-dimensional chamfer matching. Comput Aided Surg 2:346–355.
[58] Wesseling P (2004) An Introduction to Multigrid Methods. R.T. Edwards, Philadelphia.
[59] Wu T, Engelhardt M, Fieten L, Popovic A, Radermacher K (2006) Anatomically constrained deformation for design of cranial implant: Methodology and validation. In: Larsen R, Nielsen M, Sporring J (eds) 9th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), LNCS 4190, Springer, Berlin/Heidelberg, pp 9–16.
[60] Wu WZ, Zhang Y, Li H, Wang WS (2009) Fabrication of repairing skull bone defects based on the rapid prototyping. J Bioact Compat Polym 24:125–136.
[61] Yamashima T (1989) Cranioplasty with hydroxylapatite ceramic plates that can easily be trimmed during surgery. Acta Neurochir 96:149–453.