執行腦部手術前，醫師通常只能依靠術前的醫學影像(CT、MRI)、腦部解剖學與自身的手術經驗，規劃適當的手術路徑，假如規劃的手術路徑不理想，手術過程中一不小心可能會傷到腦白質神經纖維或腦血管，造成病人的手術後遺症。如果能在手術前提供醫師腦白質神經纖維與腦血管的3D影像與空間資訊，將可幫助醫師在規劃手術路徑時，預先避開腦白質神經纖維與腦血管。
本研究結合2D MRI切面影像與3D腦神經與腦血管模型，發展一套用於腦部手術的術前路徑規劃系統。研究內容包括3D腦組織的三維重建、影像對位註冊與系統使用者介面設計。在3D腦組織三維重建部分，使用FA影像重建出腦白質神經纖維結構，使用DTI影像搭配神經纖維束追蹤術，重建出腦白質神經纖維束，使用MRI影像經影像前處理後重建出顯示腦溝的腦表面模型，使用CTA影像搭配本實驗室的腦血管分割與重建方法，重建出腦血管；影像對位部分，以MRI影像的座標系作為基準，將其他影像的座標系對位到MRI影像的座標系；系統使用者介面設計部分，將2D切面影像與3D腦組織模型分別顯示於同一畫面的單獨視窗中，在視窗旁設計滑桿，透過滑桿調整畫面中手術路徑的方向。
本研究實驗內容包括3D腦組織模型評估與影像對位準確性，在3D腦組織模型的部分，可看到各種解剖特徵；在影像對位準確性的部分，FA與MRI的對位，經棋盤分類法的評估，大腦、腦灰質、腦白質與腫瘤的輪廓皆有相連，可見有足夠的準確度，CTA與MRI的對位，經測試特徵點的平均距離誤差為3.96mm，因為CTA與MRI影像來自不同的病人，實驗結果沒預期中理想。

Before performing brain surgery, most surgeons only rely on preoperative medical images (CT, MRI) and their knowledge of brain anatomy and surgical experience to plan surgical path. If the surgical path is not ideal or safe, injuring white matter nerve fibers and blood vessels during surgical procedure may occur, and thus make the patient have sequela. If 3D image information of nerve fibers and blood vessels around the lesion and surgical path are provided for the surgeon, it can assist the surgeon to plan a safe surgical path to avoid important nerve fibers and blood vessels.
This research develops a preoperative surgical path planning system which combines preoperative images with 3D reconstructed models of nerve, brain, and blood vessels for brain surgery. The research contents include 3D image reconstruction, image registration, and user interface design. The 3D image reconstruction part includes using FA and DTI images to reconstruct nerve fiber models of white matter respectively, using MRI image to reconstruct brain sulcus model, using CTA image to reconstruct brain blood vessel model. The image registration part includes the registration between MRI image frame and the FA/DTI and CTA image frames. As to the user interface design, 2D axial, sagittal and coronal slices of MRI images and 3D models of brain, white matter and blood vessels are displayed simultaneously for planning a safe surgical path.
The experiment includes the accuracy and reliability of image registrations. In image registration part, the registration accuracy between FA and MRI images is checked visually and good enough by using Checkerboard Classification Method. The average distance error of CTA and MRI image registration is 3.96mm, which is not good enough because both images are scanned from different patients.

[1] 衛生福利部. "108年國人死因統計結果." https://www.mohw.gov.tw/cp-16-54482-1.html
[2] U. Mezger, C. Jendrewski, and M. Bartels, "Navigation in surgery," (in eng), Langenbecks Arch Surg, vol. 398, no. 4, pp. 501-514, 2013
[3] R. B. Kochanski, J. M. Lombardi, J. L. Laratta, R. A. Lehman, and J. E. O’Toole, "Image-Guided Navigation and Robotics in Spine Surgery," Neurosurgery, vol. 84, no. 6, pp. 1179-1189, 2019
[4] X. Shao et al., "Virtual reality technology for teaching neurosurgery of skull base tumor," BMC Med Educ, vol. 20, no. 1, p. 3, Jan 3 2020
[5] D. A. Orringer, A. Golby, and F. Jolesz, "Neuronavigation in the surgical management of brain tumors: current and future trends," (in eng), Expert Rev Med Devices, vol. 9, no. 5, pp. 491-500, 2012
[6] Z. D. Travis, P. Sherchan, W. K. Hayes, and J. H. Zhang, "Surgically-induced brain injury: where are we now?," Chinese Neurosurgical Journal, vol. 5, no. 1, p. 29, 2019/12/16 2019
[7] Medtronic. "Stealth Station Surgical Navigation System." https://www.medtronic.com/us-en/index.html
[8] BrainLab. "Vector Vision System." https://www.brainlab.com/
[9] 馮德諼, "腦部手術導航系統與混合實境之整合," 碩士, 機械工程學系, 國立中央大學, 桃園縣, 2019.
[10] R. Wang, V. J. Wedeen, and A. A. M. C. f. B. Imaging. "TrackVis." http://trackvis.org/
[11] L. Zhukov, K. Museth, D. Breen, A. Barr, and R. Whitaker, "Level set modeling and segmentation of diffusion tensor magnetic resonance imaging brain data," J. Electronic Imaging, vol. 12, pp. 125-133, 01/01 2003
[12] W. E. Lorensen and H. E. Cline, "Marching cubes: A high resolution 3D surface construction algorithm," ACM siggraph computer graphics, vol. 21, no. 4, pp. 163-169, 1987
[13] P. Mukherjee, J. I. Berman, S. W. Chung, C. P. Hess, and R. G. Henry, "Diffusion Tensor MR Imaging and Fiber Tractography: Theoretic Underpinnings," American Journal of Neuroradiology, vol. 29, no. 4, pp. 632-641, 2008
[14] WIKIPEDIA. "Fractional anisotropy." https://en.wikipedia.org/wiki/Fractional_anisotropy
[15] S. M. Smith, G. Kindlmann, and S. Jbabdi, "Cross-subject comparison of local diffusion MRI parameters," in Diffusion MRI: Elsevier, 2014
[16] C. Thomas et al., "Anatomical accuracy of brain connections derived from diffusion MRI tractography is inherently limited," (in eng), Proc Natl Acad Sci U S A, vol. 111, no. 46, pp. 16574-16579, 2014
[17] OHBM. "Introduction to Diffusion Tractography." https://www.pathlms.com/ohbm/courses/1492/sections/1822/video_presentations/17176
[18] P. J. Besl and N. D. McKay, "A method for registration of 3-D shapes," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 14, no. 2, pp. 239-256, 1992
[19] K. Rohr, H. S. Stiehl, R. Sprengel, T. M. Buzug, J. Weese, and M. H. Kuhn, "Landmark-based elastic registration using approximating thin-plate splines," IEEE Transactions on Medical Imaging, vol. 20, no. 6, pp. 526-534, 2001
[20] B. J. Jellison, A. S. Field, J. Medow, M. Lazar, M. S. Salamat, and A. L. Alexander, "Diffusion Tensor Imaging of Cerebral White Matter: A Pictorial Review of Physics, Fiber Tract Anatomy, and Tumor Imaging Patterns," American Journal of Neuroradiology, vol. 25, no. 3, pp. 356-369, 2004
[21] WIKIPEDIA. "Corpus callosum." https://en.wikipedia.org/wiki/Corpus_callosum
[22] WIKIPEDIA. "Cingulum " https://en.wikipedia.org/wiki/Cingulum_(brain)
[23] WIKIPEDIA. "Corticospinal tract." https://en.wikipedia.org/wiki/Corticospinal_tract
[24] WIKIPEDIA. "Longitudinal fissure." https://en.wikipedia.org/wiki/Longitudinal_fissure
[25] WIKIPEDIA. "Central sulcus." https://en.wikipedia.org/wiki/Central_sulcus
[26] WIKIPEDIA. "Lateral sulcus." https://en.wikipedia.org/wiki/Lateral_sulcus
[27] M. Malinsky, R. Peter, E. Hodneland, A. J. Lundervold, A. Lundervold, and J. Jan, "Registration of FA and T1-weighted MRI data of healthy human brain based on template matching and normalized cross-correlation," Journal of digital imaging, vol. 26, no. 4, pp. 774-785, 2013