臺灣博碩士論文加值系統

本研究針對臨床醫療於在線適應性放射治療方面兩種情況，其一為拍攝時間不同所造成的姿態不一樣及機台不同所造成的大小不一樣的問題；另一則為病人在療程中體型或腫瘤產生顯著變化，使得體內腫瘤位置、器官相對位置與原定放射治療規劃不同。分別針對此兩種情況建立影像匹配演算法、撰寫自動化分析流程並討論匹配後之結果。
關於姿態不一樣及影像大小不一樣的部分，影像匹配步驟依序如下：利用影像中骨頭與其他組織間具有HU值差異性之特性，對電腦斷層掃描影像以及錐形束電腦斷層掃描影像進行二值化處理，分別獲得兩種影像的萃取骨頭之影像；接著通過仿射匹配演算法進行二者影像間對位、旋轉、縮放之影像匹配，並建立兩張影像所對應之仿射矩陣；最後將仿射矩陣代入電腦斷層掃描影像圈選之各器官影像，以進行仿射變形運算得到變形之器官輪廓。
針對體內腫瘤位置、器官相對位置與原定放射治療規劃不同的情形，匹配步驟依序如下：先針對變形之器官輪廓以及錐形束電腦斷層掃描影像進行特徵匹配，以獲得兩者相對應之特徵點；再利用器官輪廓特徵點，針對變形之器官輪廓進行薄板樣條插值，以得到錐形束電腦斷層掃描影像之器官輪廓。
此研究亦將所提出之剛性匹配與非剛行匹配演算法，皆以Python撰寫半自動化分析流程，並分別針對肺、心臟進行測試，皆可順利完成器官輪廓匹配。

This study addresses two scenarios of clinical treatment in the field of adaptive radiotherapy. One is the problem of different postures and sizes caused by different shooting times and different machines; the other is when the patient’s body size or tumor changes significantly during the course of treatment, resulting in a different position of the tumor and the relative position of organs in the body than the original radiation treatment plan. These two situations will be addressed and discussed separately.
The posture and size differences are handled as : The computed tomography image and the cone beam computed tomography image are binarized to obtain the extracted bone images of the two images using the feature of the difference in HU values between bone and other tissues in the images, and then the affine matching algorithm is used to align, rotate and scale the two images to establish an image match between the two images. Then, the image matching algorithm is used to match the image alignment, rotation and scaling between the two images and create the affine matrix of the two images, and then the affine matrix is substituted into the images of each organ circled by the computed tomography images to obtain the deformed organ contours.
The processing progress for the relative positions of tumor location and organs different from the original radiotherapy protocol is as follows: the deformed organ contours are feature matched with the cone beam computed tomography images, and finally the deformed organ contours are interpolated with thin-plate samples to obtain the organ contours of the cone beam computed tomography images.
The proposed image registration algorithm included rigid and non-rigid registration would be performance by the self-written Python program. The image registration cases of lungs and heart had been adopted to test the semi-automation processing, and it has good performance of contours deformed from CT for radiation therapy planning to CBCT for online therapy.

摘要 i
Abstract ii
誌謝 iii
目錄 iv
圖目錄 vi
符號表 ix
第一章 緒論 1
1.1 研究背景與動機 1
1.2 文獻回顧 3
1.3 論文架構 5
第二章 剛性匹配 7
2.1 影像介紹 7
2.2 器官萃取 12
2.3 剛性匹配 15
2.3.1仿射匹配 15
2.3.2仿射變形 17
2.4 結果與討論 18
第三章 非剛性匹配 19
3.1 3D模型重建 19
3.2 特徵匹配 23
3.2.1高效率的特徵點檢測演算法FAST 24
3.2.2二進位強健獨立特徵點描述演算法BRIEF 26
3.2.3結果與討論 27
3.3 點群應用 29
3.3.1薄板樣條插值原理 29
3.3.2結果與討論 31
3.4 結果與討論 32
第四章 案例分析 34
第五章 結論與未來規劃 41
5.1 結論 41
5.2 未來規劃 42
參考文獻 43

[1]中華民國衛生福利部，110年死因統計結果分析，2022
[2]ICRU Report 42. Use of computers in external beam radiotherapy procedures with high energy photons and electrons. Washington, DC: International Commission on Radiation Unit and Measurement, 1988.
[3]D. Yan, F. Vicini, J. Wong and A. Martinez, “Adaptive radiation therapy,” Physics in Medicine & Biology, vol. 42, no.1, pp. 123-132, 1997.
[4]W. Q. Jackie, L. Taoran, W. Qiuwen, Y. Fang-Fang, “Adaptive Radiation Therapy Technical Components and Clinical Applications,” Cancer Journal, vol. 17, no.3, pp. 182-189, 2011.
[5]S. Lim-Reinders, B. M. Keller, S. Al-Ward, A. Sahgal, A. Kim, “Online Adaptive Radiation Therapy,” International Journal of Radiation Oncology·Biology·Physics, vol. 99, no. 4, pp. 994-1003, 2017.
[6]S. Acharya, B. W. Fischer Valuck, R. Kashani, P. Parikh, D. Yang, T. Zhao, O. Green, O. Wooten, H. H. Li, Y. Hu, V. Rodriguez, L. Olsen, C. Robinson, J. Michalski, S. Mutic, J. Olsen, “Online Magnetic Resonance Image Guided Adaptive Radiation Therapy: First Clinical Applications,” International Journal of Radiation Oncology·Biology·Physics, vol. 94, no. 2, pp. 394-403, 2016.
[7]S. Trebeschi, J. J. M. van Griethuysen, D. M. J. Lambregts, M. J. Lahaye, C. Parmer, F. C. H. Bakers, N. H. G. M. Peters, R. G. H. Beets-Tan, H. J. W. L. Aerts, “Deep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR,” Scientific Reports, vol. 7, no. 5301, 2017.
[8]P. Meyer, V. Noblet, C. Mazzara, A. Lallement, “Survey on deep learning for radiotherapy,” Computers in Biology and Medicine, vol. 98, no. 1, pp. 126-146, 2018.
[9]G. Subsol, N. Roberts, M. Doran, J. P. Thirion and G. H. Whitehouse, “Automatic analysis of cerebral atrophy,” Magnetic Resonance Imaging, vol. 15, no. 8, pp. 917-927, 1997.
[10]H. Wang, L. Dong, J. O’Daniel, R. Mohan, A. S. Garden, K. K. Ang, D. A. Kuban, M. Bonnen, J. Y. Chang and R. Cheung, “ Validation of an accelerated ‘demons’ algorithm for deformable image registration in radiation therapy,” Physics in Medicine and Biology, vol. 50, no. 12, pp. 2887-2905, 2005.
[11]Y. Z. Szeto, M. G. Witte, M. van Herk and J.-J. Sonke, “A population based statistical model for daily geometric variations in the thorax,” Radiotherapy and Oncology, vol. 123, no. 1, pp. 99-105, 2017.
[12]D. L. G. Hill, D. J. Hawkes, N. A. Harrison, C. F. Ruff, “A strategy for automated multimodality image registration incorporating anatomical knowledge and imager characteristics,” Information Processing in Medical Imaging, vol. 687, pp. 182-196, 1993.
[13]P. Elsen, E. Pol, T. Sumanaweera, P. Hemler, S. Napel, J. Adler, “Grey value correlation techniques used for automatic matching of CT and MR brain and spine images,” Visualization in Biomedical Computing 1994, vol. 2359, pp. 227-237, 1994.
[14]F. P. M. Oliveira, J. M. R. S. Tavares, “Medical image registration: a review,” Computer Methods in Biomechanics and Biomedical Engineering, vol. 17, no. 2, pp. 79-93, 2014.
[15]D. Sengupta, P. Gupta, A. Biswas, “A survey on mutual information based medical image registration algorithms,” Neurocomputing, vol. 486, pp. 174-188, 2022.
[16]C. Paganelli, G. Meschini, S. Molinelli, M. Riboldi, G. Baroni, “Patient-specific validation of deformable image registration in radiation therapy: overview and caveats,” Medical Physics, vol. 45, no. 10, pp. e908-e922, 2018.
[17]F. B. Zitova and J. Flusser, “Image registration methods:a survey,” Image and Vision Computing, vol. 21, no. 11, pp. 997-1000, 2013.
[18]J. P. W. Pluim, J. B. A. Maintz and M. A. Viergever, “Mutual-information-based registration of medical images: a survey,” IEEE Transactions on Nuclear Science, vol. 22, no. 8, pp. 986-1004, 2003.
[19]F. Maes, D. Vandermeulen and P. Suetens, “Medical image registration using mutual information,” Proceedings of the IEEE, vol. 91, no. 10, pp. 1699-1722, 2003.
[20]D. G. Lowe, “Distinctive Image Features from Scale-Invariant Keypoints,” International Journal of Computer Vision volume, vol. 60, pp. 91-110, 2004.
[21]E. Rublee, V. Rabaud, K. Konolige, G. Bradski, “ORB: An efficient alternative to SIFT or SURF,” International Conference on Computer Vision(ICIV 2011), November 06-13, 2011, Barcelona, Spain.
[22]E. Rosten, T. Drummond, “Machine Learning for High-Speed Corner Detection”, The 9th European Conference on Computer Vision(ECCV 2006), May 07-13, 2006, Graz, Austria.
[23]M. Calonder, V. Lepetit, C. Strecha, P. Fua, “BRIEF: Binary Robust Independent Elementary Features,” The 11th European Conference on Computer Vision(ECCV 2010), September 05-11, 2010, Heraklion, Crete, Greece.
[24]C. R. Meyer, J. L. Boes, B. Kim, P. H. Bland, K. R. Zasadny, P. V. Kison, K. Koral, K. A. Frey, R. L. Wahl, “Demonstration of accuracy and clinical versatility of mutual information for automatic multimodality image fusion using affine and thin-plate spline warped geometric deformations,” Medical Image Analysis, vol. 1, no. 3, pp. 195-206, 1997.
[25]A. Dini and E. Rahtu, “TPSAD: Learning to Detect and Localize Anomalies With Thin Plate Spline Transformation,” The 26th International Conference on Pattem Recognition(ICPR 2022), August 21-25, 2022,Montreal, QC, Canada.
[26]N. S. Vadivu, G. Gupta, Q. N. Naveed, T. Rasheed, S. K. Singh, D. Dhabliva, “Correlation-Based Mutual Information Model for Analysis of Lung Cancer CT Image”, BioMed Research International, vol. 22, Article ID 6451770,12 pages, 2022.
[27]M. M. Ter-Pogossian, E. S. Weiss, R. E. Coleman and B. E. Sobel, “Computed tomography of the heart,” American Journal of Roentgenology, vol. 127, no. 1, 1976.
[28]G. Treglia, B. Muoio, L. Giovanella, M. Salvatori, “The role of positron emission tomography and positron emission tomography/computed tomography in thyroid tumours: an overview,” European Archives of Oto-Rhino-Laryngology, vol. 270, pp. 1783-1787, 2013.
[29]I. Sluimer, A. Schilham, M. Prokop, B. V. Ginneken, “Computer analysis of computed tomography scans of the lung: a survey,” IEEE Transactions on Medical Imaging, vol. 25, pp. 385-405, 2006.
[30]M. P. Federle, H. I. Goldberg, J. A. Kaiser, A. A. Moss, R. Ｂ. Jeffrey, J. C. Mall, “Evaluation of abdominal trauma by computed tomography,” Radiology, vol. 138, no. 3,1981.
[31]Cone beam vs.fan beam CT,
https://surgeryreference.aofoundation.org/cmf/further-reading/cas-cone-beam-vs-fan-beam-ct
[32]W. E. Loernsen, H. E. Cline, “Marching cubes: A high resolution 3D surface construction algorithm,” ACM SIGGRAPH Computer Graphics, vol. 21, no. 4, pp. 163-169, 1987.
[33]Cong-Kai Lin, Eileen Chih-Ying Yang, “4D-CT-Based Pulmonary Ventilation Analysis Mapping Automation”, The 19th International Conference on Automation Rechnology(Automation 2022), November 11-13, 2022, Kaohsiung, Taiwan.
[34]E. Rosten and T. Drummond, “Machine Learning for High-Speed Corner Detection,” Computer Vision – ECCV 2006, pp. 430–443, 2006.
[35]M. Calonder, V. Lepetit, C. Strecha, and P. Fua1, “BRIEF: Binary Robust Independent Elementary Feature,” Computer Vision – ECCV 2010, pp. 778–792, 2010.
[36]Thin-plate smoothingbspline.
https://www.mathworks.com/help/curvefit/tpaps.html