跳到主要內容

臺灣博碩士論文加值系統

(54.91.62.236) 您好!臺灣時間:2022/01/17 23:44
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:徐榮祥
研究生(外文):Hsu jung hsiang
論文名稱:腫瘤偵測與顱顏骨骼重建
論文名稱(外文):Tumor Detection and Craniofacial Implant Reconstruction
指導教授:曾清秀曾清秀引用關係
指導教授(外文):Tseng C. S.
學位類別:博士
校院名稱:國立中央大學
系所名稱:機械工程研究所
學門:工程學門
學類:機械工程學類
論文種類:學術論文
論文出版年:2002
畢業學年度:90
語文別:英文
論文頁數:96
中文關鍵詞:腫瘤邊界偵測顱顏重建超音波影像
外文關鍵詞:TumorCraniofacialBoundary detectionrapid protyping machine
相關次數:
  • 被引用被引用:3
  • 點閱點閱:319
  • 評分評分:
  • 下載下載:72
  • 收藏至我的研究室書目清單書目收藏:0
本論文題出一個新的顱顏重建的方法解決重建的問題降低手術所須要的時間,以類神經網路預測病灶區的外形並於臨床應用獲得良好的結果
另外本論文亦提出方有效的超音波偵測乳房腫瘤外型並用於腫瘤良惡性的判斷
Traditionally, plastic surgeons reconstruct craniofacial defects according to their clinic experience while operation is in progress. It is time-consuming to make the implant and the hand-made implant is usually difficult to well match the defect. The purpose of this research is to propose a method to improve the effect and efficiency of traditional operation. In this study, the orthogonal neural network is applied to predict the surface model of the defect and then the Marching Cube Algorithm is applied to reconstruct the 3D defect implant model. A rapid prototyping machine can accurately produce the geometric of patient-specific implants in acrylic resin. Two clinic cases with either a forehead defect or a large skull defect are given to evaluate the performance of the proposed method. The results show that the reconstructed implants fit into the defects well .
A method for tumor boundary detection and a procedure for the diagnosis of breast tumor are also presented. The grey level projection distribution of the ROI is adopted to determine the seed point and threshold value of the tumor. Then the tumor boundary can be determined by searching from the seed point and by using the region growth method. After the tumor boundary of each image slice has been determined, the tumor size and spatial position can be calculated accurately. The shape and margin of the detected tumor boundary can also be used to assist the prediction of breast tumor attributes. The method has been applied to detect the breast tumor boundary from sonograms and brain tumor boundary from CT image slices. The results of clinic tests show that the computer generated tumor boundary matches well with the subjective judgement of an experienced breast tumor expert and a neurosurgeon.
In this study, fifty-four breast sonograms are analysed. In comparison with physician judgement, twenty-three cases reach 100% similarity. Fifteen cases reach 90% similarity and eleven cases reach 80%. However, one case only reaches 70% and four cases are different from the physician judgement.
Contents
AbstractⅠ
ContentsⅢ
List of figures and tablesⅤ
Chapter Ⅰ: Introduction
1.1 Research Motivation1
1.2 Literature review2
1.3 Method6
1.4 Capsule summary8
Chapter Ⅱ: Boundary Detection of Bone Defects and Tumors
2.1 Defect boundary detection from CT images10
2.2 Tumor boundary detection from breast sonogram13
2.3 Boundary detection of brain tumor27
2.4 Implant boundary predicted by orthogonal neural network
30
Chapter Ⅲ: Reconstruction of Craniomaxillary Defects
3.1 Traditional defect reconstruction42
3.2 Surface prediction by orthogonal neural networks43
3.3 Malocclusion adjustment for mandible reconstruction51
Chapter Ⅳ: Boundary Detection of Ultrasound Images for the Diagnosis of Breast Tumor
4.1 Sonographic feature for Breast tumors58
4.2 Distinction between benign and malignant breast lesion61
4.3 Discussion69
Chapter Ⅴ: Boundary Detection and Reconstruction of Brain Tumor
5.1 Boundary detection of brain tumor from CT images------71
5.2 Reconstruction of brain tumor74
5.3 Discussion78
Chapter Ⅵ :Conclusion 80
References83
Appendix 88
List of figures and tables
Figure 1-1 An ultrasound image with carcinomal tumor4
Figure 1-2 The tumor boundary detected by the Sobel method4
Figure 1-3 Boundary detection using the Snake-Balloon method6
Figure 1-4 The breast tumor with blood vessel around and fat inside6
Figure 2-1 Boundary detection of a skull CT image with a defect 12
Figure 2-2 Boundary detection of a skull CT image without defect 13
Figure 2-3 The flowchart of the breast tumor boundary detection 14
Figure 2-4 Selection of the ROI15
Figure 2-5 Grey level projection distribution in horizontal direction15
Figure 2-6 Grey level projection distribution in vertical direction16
Figure 2-7 Distance distribution of boundary points19
Figure 2-8 Procedure for tumor shape determination 21
Figure 2-9 Procedure for tumor margin determination22
Figure 2-10 The change of ROI vs. standard deviation26
Figure 2-11 The procedure of brain tumor boundary detection31
Figure 2-12 A head CT image with brain tumor32
Figure 2-13 The brain tumor boundary32
Figure 2-14 A typical structure of the orthogonal neural network33
Figure 2-15 Implant surface prediction by the orthogonal neural network36
Figure 2-16 Selection of the bone boundary around the defect37
Figure 2-17 The predicted boundary curve has a larger curvature37
Figure 2-18 The predicted boundary curve has a smaller curvature38
Figure 2-19The predicted curves based on the two selected bone boundaries38
Figure 2-20 Surface prediction procedure by 3D orthogonal neural network40
Figure 2-21 The construction of a 3D orthogonal neural network41
Figure 3-1 The defect crosses the central symmetric plane43
Figure 3-2 One of the head CT slices44
Figure 3-3 The predicted outer boundary curve of the defect44
Figure 3-4 The defect area marked by the region growth method45
Figure 3-5 The generated implant model45
Figure 3-6 The generated implant model fits the defect perfectly46
Figure 3-7 The real implant is fitted into the defect46
Figure 3-8 The bone coordinates around the defect derived by the Sobel method (in black) vs. by the neural network (in grey)47
Figure 3-9 A large defect on the left side of the skull49
Figure 3-10 The reconstructed implant is fitted into the defect49
Figure 3-11 The reconstructed implant model50
Figure 3-12(a) The reconstructed implant by 4x4 Lendegre polynomials52
Figure 3-12(b) The reconstructed implant by 3x3 Lendegre polynomials52
Figure 3-13 The patient suffered from malocclusion53
Figure 3-14 The mandibular segments are fixed by a fixation plate53
Figure 3-15 The 3D mandible prior to surgery53
Figure 3-16 The left residual mandible54
Figure 3-17 The right residual mandible55
Figure 3-18 The two residual mandibles (in red) are adjusted to their original positions55
Figure 3-19 A cutting plane cuts the right mandible segment from the normal mandible56
Figure 3-20 The fixation plate is bent along with the mandibular model57
Figure 4-1 Definition of quadrants63
Figure 4-2 The breast tumor shape marked by the surgeon66
Figure 4-3 The breast tumor shape generated by the proposed method66
Figure 4-4 The breast tumor shape marked by the surgeon66
Figure 4-5 The breast tumor shape generated by the computer67
Figure 4-6 The tumor is surrounded by fat 67
Figure 4-7 The tumor boundary generated by the proposed method68
Figure 4-8 Classification of diagnosis based on tumor shape and margin 70
Figure 5-1 The original CT image with brain tumor73
Figure 5-2 The CT image after Histogram-equalized enhancement 74
Figure 5-3 The blood block region 74
Figure 5-4 The hand-drawn tumor contour75
Figure 5-5 List of boundary detection results76
Figure 5-6 The 3D model of the skull and brain tumor 79
Table 2-1 Correlation ratio for different Ks and Kc28
Table 4-1 The list of tumor margins and shapes63
Table 4-2 Tumor shape/margin vs. biopsy result69
Appendix 188
Appendix 291
References:
[1]Aboutanos, G.B., Nikanne, J. N. Watkins, and Dawant, B. M., “Model Creation and Deformation for the Automatic Segmentation of the Brain in MR Images,” IEEE Transactions on biomedical engineering, Vol. 46, No. 11, pp.145-160 1999.
[2]Ackerman, M.J., “The visible human project,” Journal of Bio-communication, Vol. 10, pp.14, 1991.
[3]Adams, R. and Bischof, L., “Seeded region growing,” IEEE Trans. Pattern Anal. Machine Intelligent, Vol.16, pp. 641-647, 1994.
[4]Bamber, J.C., Gonzales, L.D., Cosgrove, D.O. et al., “Quantitative evaluation of real-time ultrasound features of the breast,” Ultrasound Medical Biological, Vol. 14, pp. 59, 1988.
[5]Baxes, G.A. Digital Image Processing principles and Application. John Willy & Sons (USA), pp 217-235, 1994.
[6]Chalana, V. and Kim, Y., “A Methodology for Evaluation of Boundary Detection Algorithms On Medical Image,” IEEE Transactions on biomedical engineering, Vol. 16, No. 5, pp. 110-127, 1997.
[7]Chang, S.C.N., Liao, Y.F., Hung, L.M., Tseng, C.S., Hsu, J.H., and Chen, J.K., “Prefabricated Implants or Grafts with Reverse Models of Three-Dimensional Mirror-Image Templates for Reconstruction of Craniofacial Abnormalities,” Plastic and Reconstructive Surgery, Vol. 104, No. 5, pp. 1413-1418, 1999.
[8]Chao, T.C., Lo, Y.F., Chen, S.C., and Chen, M.F., “Prospective sonographic study of 3093 breast tumors,” Journal of Ultrasound Medicine, Vol.18, pp.363-370, 1999.
[9]Chen, M.S. and Manry, M.T., “Conventional modelling of the multiplayer perception using polynomial basis function,” IEEE Transactions Neural Networks, Vol. 4, pp.164-166, 1993.
[10]Elsen, P. A. van den, Pol, E. J. D., and Viergever, M. A., “Medical image matching— a review with classification,” IEEE Engineering in medicine and biology, Vol.12, pp.26—39, 1993.
[11]Fajardo, L., Hillman, B.J., and Fery, C., “Correlation between breast parenchymal patterns and mammography certainty of diagnosis,” Invert Radio, Vol.23, pp.505, 1998.
[12]Gall, K.P. and Verhey, L.J., “Computer-assisted positioning of radiotherapy patients using implanted radio-opaque fiducials,” K. P. Medical physics, Vol. 20, pp.1152—1159,1993.
[13]Garra, B.S. and Krasner, B.H. et al., “Improving the distinction between benign and malignant breast lesions the value of sonographic texture analysis,” Journal Ultrasound Medicine, Vol. 13, pp.267-285, 1993.
[14]Gonzalez, R.C. and Woods, R.E. Digital image processing. Addison-Wesley (USA), pp. 413-431, 1993.
[15]Gonzalez, R.C. and Woods, R.E. Digital image processing. Addison-Wesley (U.S.A), pp458-465, 1993.
[16]Hall, F.M., “Sonography of the breast Controversies and options,” American Journal of Roentgenology, Vol.169, pp.1635-1647, 1997.
[17]Haralick, R.M. and Shapiro, L.G. Computer and Robot Vision Reading. Addison Wesley (U.S.A), pp.482-490, 1992.
[18]Harder, R.L., “Interpolation using surface alpine,” Journal Aircraft, Vol. 9 No. 2, pp. 189-191, 1972.
[19]Hildebrand, F. B. Advanced Calculus for Application. Prentice-Hall Inc. (USA), pp.208-221, 1976.
[20]Howe, R.D., Matsuoka, Y., “Robotics for Surgery,” AnnualReview Biomedical Engineering, Vol. 1, pp211-240, 1999.
[21]Hsu, J.H. and Tseng, C.S., “A Methodology for Evaluation of Boundary Detection Algorithms on Breast Ultrasound Images,” Journal of Medical Engineering & Technology (MET 100427,submitted 28 June 2001).
[22]Hsu, J.H. and Tseng, C.S., “Application of Three-Dimensional orthogonal neural network to craniomaxillary reconstruction,” Computerized Medical imaging and graphics (CMIG 434, submitted 25 May 2001).
[23]Hsu, J.H. and Tseng, C.S.,“Application of orthogonal neural network to craniomaxillary reconstruction,” Journal of Medical Engineering & Technology, Vol. 24, No. 6, pp. 262-266, 2000.
[24]Huang, H.P. and Chen, J.J., “Unstable Back propagation Method in Neural Networks,” International Conference on Automatic Technology, Taiwan 1, pp.605-618, 1990.
[25]Jonathan, C.C., Fright, W. R., and Beatson, R. K., “ Surface Interpolation with Radial Basis Function for Medical Imaging,” IEEE Transaction Medical Imaging, Vol. 16, pp. 96-107, 1997.
[26]Kass, M., Witkin, A., and Terzopoulos, D., “Snakes: Active contour models,” in Proc. 1st International Conference Computer Vision, London, pp.259-268, 1987.
[27]Kasumi, F., “Can microcalcification located within breast carcinomas be detected by ultrasound imaging,” Ultrasound Medical Biological, Vol.14 pp.175, 1988.
[28]Kolb, T.M., Lichy, J., and Newhouse, J.H., “Occult cancer in women with dense breasts: Detection with screening US-Diagnostic yield and tumor characteristics,” Radiology, Vol.207, pp.191-202, 1998.
[29]Lam, K.L., Haken, R.K.T., McShan, D.L., and Thornton, A.F., “Automated determination of patient setup errors in radiation therapy using spherical radio-opaque markers,” Medical physics, Vol. 20, No.4, pp.1145—1152, 1993.
[30]Linney, A.D. and Tan, A.C. et al., “Three-dimensional visualization of data on human anatomy: Diagnosis and surgical planning,” Journal Audiovisual Media in Medical, Vol. 16, pp. 4-10, 1993.
[31]Lunsford, L.D. Modern stereotactic neurosurgery. Martinis Nijhoff (U.S.A), pp.651-659, 1988.
[32]Ma, L., Fishell, E., Wright, B. et al., “Case control study of factors associated with failure to detect breast cancer by mammography,” Journal Natl. Cancer Inst, Vol.84, pp.781-790, 1992.
[33]Mai, C.C., Tseng, C.S., Chen, C.S., ”An Orthogonal Neural Controller with The Application in a Ball System Control,” International Journal of Intelligent Control and Systems, Vol.3, No. 2, pp. 223-236, 1999.
[34]Qian, S., Lee, Y.C., Jones, R.D., Barnes, C. W., and Lee, K., “Function Approximation with an Orthogonal Basis Net,” International Journal of Computer Neural Network, Vol. 3, pp. 605-618, 1990.
[35]Schroeder, W.S., Martin, K., and Loren son, B. The Visualization Toolkit. Prentice-Hall Inc. (USA), pp. 83-92, 1997.
[36]Skaane, P. and Engedal, K., “Analysis of sonographic feature in the differentiation of fibro-adenoma and invasive ducal carcinoma,” American Journal of Roentgenology, No. 170, pp.109-114, 1998.
[37]Sonka, M., Hlavac, V., and Boyle, R. Image Processing, Analysis and Machine Vision. Chapman and Hall (U.K.), pp.512-519, 1993.
[38]Staren, E.D. and Oneill, T.P., “Breast ultrasound,” Surgery Clinical North American, Vol.78, pp.219-232, 1988.
[39]Ueno, E., Tohno, E., Soeda, S. et al., “Dynamic tests in real-time breast echography,” Ultrasound Medical Biological, Vol. 14 pp.53 1998.
[40]Venta, L.A., Dudiak, C.M., and Salomen, C. G., “Sonographic evaluation of the breast,” Radiographic, Vol. 14 pp. 29, 1994.
[41]VP, J., “The role of US in breast imaging,” Radiology, Vol. 177, pp.305-311, 1990.
[42]Watt, A. 3D Computer Graphics. Addison Wesley (U.S.A), pp.163-221, 1993.
[43]Yang, S.S. and Tseng, C.S., “An Orthogonal Neural Networks for Functional Approximation,” IEEE Transactions on systems, man, and cybernetics, Vol. 26, pp.779-785, 1996.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
1. 47. 蔡嬪嬪,1992,「固態離子導體電化學式氣體感測器」,材料與社會,第68期,頁67-70。
2. 42. 曾明漢,1992,「觸媒燃燒型氣體感測器」,材料與社會,第68期,頁57-61。
3. 26. 吳世雄,「侷限空間安全-災害預防對策」,勞工安全衛生簡訊第4期。Pgs.11.
4. 41. 湯大同,「再談地下通風不良場所之危險性」,勞工安全衛生簡訊第17期。Pgs.3.
5. 33. 李仙吉、林賴金、葉光明、謝光明、邱昭正、張嘉雄,1991,「意外傷害之預防」,健康教育,67期,頁48-53。
6. 32. 黃國珍,1992,「意外事故的模型與分析」,警專學報,頁431-449。
7. 3. 李慶恩,白炳豐,訂單需求導向之零工型生產產能批量規劃方法,管理與系統,第三卷,第一期,民國八十五年一月,pp.43-62.。
8. 51. 吳仁彰,2000,「固態電解質式一氧化碳感測器之理論研究」,量測資訊,第73期,頁44-47。
9. 45. 邱秋燕、周澤川,1993,「化學感測器之原理與應用」,化工,第40卷,第3期,頁120-133。
10. 44. 陳一誠,1992,「金屬氧化物半導體型氣體感測器」,材料與社會,第68期,頁62-66。
11. 43. 顧志鴻,1992,「MOSFET氣體感測器」,材料與社會,第68期,頁71-77。
12. 49. 吳仁彰,1998,「氣體分析儀和氣體感測器之校正及不確定評估」,量測資訊,第60期,頁64-70。
13. 50. 鄭錦榮,1997,「伽伐尼腐蝕的電化學原理」,塗料與塗裝技術,第62期,頁47-56。
14. 17. 黃金銀,1998,「電信業人孔作業勞工一氧化碳暴露之研究」,中國文化大學勞工研究所碩士論文。
15. 12. 傅還然,1994,「台灣地區侷限空間重大災害報告」,工業安全科技,第15期。