(3.238.186.43) 您好!臺灣時間:2021/02/28 12:15
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:::

我願授權國圖
: 
twitterline
研究生:黃嘉彥
研究生(外文):Jia-Yann Huang
論文名稱:核子醫學骨骼掃描自動骨骼病灶位置偵測
論文名稱(外文):Automatic lesions detection of nuclear medicine bone scan
指導教授:陳永盛陳永盛引用關係
指導教授(外文):Yung-Sheng Chen
學位類別:碩士
校院名稱:元智大學
系所名稱:電機工程學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2004
畢業學年度:92
語文別:英文
論文頁數:114
中文關鍵詞:99mTc-MDP全身骨骼掃描灰階值影像分割病灶自動偵測
外文關鍵詞:99mTc-MDP whole body bone scangray scaleimage segmentationautomatic lesion detection
相關次數:
  • 被引用被引用:0
  • 點閱點閱:247
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
核醫全身骨骼掃描主要運用在偵測癌症骨轉移及骨髓感染等,病人在靜脈注射99mTc MDP放射藥物約3小時後,再經技術師操作伽傌閃爍攝影機為病人做全身骨骼掃描,掃描結果影像先以灰階形態儲存於電腦硬碟內,隨後經技術師調整影像亮度及對比,再經影像曝光機及洗片機將全身骨骼影像呈現在感光膠片上。臨床上核子醫學專科醫師經由視覺判讀感光膠片上骨骼影像是否呈現異常灰階升高位置,來進行骨骼病灶的診斷。但由於不同技術師之間的設備操作習慣、影像曝光機和洗片機設定參數的改變都會影響最後感光膠片上影像的品質及灰階值,因此在小病灶的發現及追蹤檢查上會造成醫師診斷上的困擾。
本論文針對臨床上核子醫學專科醫師的上述困擾,將電腦硬碟內的原始全身骨骼掃描影像資料經影像處理後,自動偵測病灶大略位置、大小,以提供核子醫學專科醫師參考。本論文期望藉由提供自動偵測結果,能輔助核子醫學專科醫師,與經由視覺判讀感光膠片上骨骼影像的結果做比對和確認,提高正確診斷率,造福病患。
本論文發展一套影像處理程式,應用型態學(morphology)、模糊集合閥值選定[ 10] 、邊緣偵測及以解剖學為基礎的影像分割法;經由模糊閥值切割影像灰階分佈,消除軟組織影像而保留寬、厚骨骼的影像,以利於解剖位置參考點的定位,進而將同質性骨骼分割出來。由於全身各部位的骨骼質量不同造成99mTc MDP放射藥物吸收率也不同,其灰階分佈寬度也不同,我們統計分析一百個樣本,分析病灶閥值和灰階分佈寬度及標準差(standard deviation)的相關性並得到一個計算病灶閥值的公式,做為最後自動判斷病灶的基準。
本論文以量化方式輔助核子醫學專科醫師做全身骨骼掃描診斷,也可能進一步應用在未來發展中的其他核子醫學全身掃描資訊,以發展更新的臨床診斷應用。

The main applications of nuclear medicine whole body bone scanning include detection of malignancy with bone metastasis and bone infection such as osteomyelitis. At 3 hours after intravenous injection 99mTc MDP, the whole body bone scan is acquired by a scintillation gamma camera equipped with a computer. The radiology technologists will adjust the brightness and contrast of whole body bone images, and then make a hard copy through formatter and film processor. Nuclear medicine physicians make a diagnosis of the study based on inspection upon the film to find out high activity lesions. Because different technologists will set different parameters in formatter and film processor, the non-standardized film display format can always confuse physicians in identifying small bone lesions and determining subtle bone lesion changes in sequential studies.
Due to the above mentioned problems, we try to use computer-aided diagnosis system to analyze and quantify the original image directly from the hard disk of gamma camera, then processing the image data for identifying the possible location and size of a lesion. We expect such an automatic lesion detection program may help physicians to identifying potential bone lesions, further increasing the diagnostic accuracy.
In this thesis, we develop an image process program based on fuzzy sets histogram thresholding method[10], edge detection method and anatomical knowledge-based image segmentation method. In order to locate the anatomical reference points, the fuzzy sets histogram thresholding and edge detection method are adopted to suppress the soft tissue and to reserve the portion of thick/wide bones in a whole body bone scan image. Next, anatomical knowledge-based image segmentation method will be applied to segment the skeletal frame into different regions of homogeneous bones. For different segmented bone regions, the lesion threshold will be set at different cut-offs in skeleton sections. We probed into the correlation of lesion threshold, gray-level distribution range and standard deviation in 100 samples, and obtained a formula of lesion threshold based on gray-level distribution range.
We expect that the results in this thesis can be applied to help nuclear medicine physicians in identifying bone lesions, and to encourage developments of other new clinical applications in nuclear medicine field.

摘要 Ⅰ
ABSTRACT Ⅱ
ACKNOWLEDGMENTS Ⅳ
CONTENTS Ⅴ
LIST OF TABLES Ⅶ
LIST OF FIGURES Ⅵ
CHAPTER 1 INTRODUCTION 1
1.1 GAMMA CAMERA 1
1.2 NUCLEAR MEDICINE BONE IMAGING 3
1.3 MOTIVATION OF OUR RESEARCH 7
1.4 OUTLINE OF THESIS 9
CHAPTER 2 BASIC FRAMEWORK OF A WHOLE BODY BONE SCAN IMAGE 12
CHAPTER 3 PRE-PROCESSING 30
3.1 THRESHOLDING AT THE VALLEY OF HISTOGRAM 31
3.2 MORPHOLOGY EROSION 32
3.3 DETERMINATION THE TRUE SIZE OF A WHOLE BODY BONE SCAN IMAGE 33
CHAPTER 4 SEGMENTATION 37
4.1 HISTOGRAM THRESHOLDING USING FUZZY SETS 37
4.2 EDGE DETECTION 46
4.3 LOCATION REFERENCE POINTS 48
4.4 HEAD SEGMENTATION 56
4.5 ARMS SEGMENTATION 60
4.6 PELVIS SEGMENTATION 66
4.7 SPINE SEGMENTATION 69
4.8 LEGS SEGMENTATION 73
4.9 CHEST SEGMENTATION 77
CHAPTER 5 ANALYSIS 80
CHAPTER 6 EXPERIMENTS AND DISCUSSIONS 91
CHAPTER 7 PERFORMANCE EVAULATION 106
CHAPTER 8 CONCLUSION 109
REFERENCES 113

[1] A. HO. Scintillation camera with multichannel collimators. J Nucl Med 1964; 5:515-531.
[2] A. HO, Resenthall DJ. Scintillation camera and position camera. In Medical radioisotope scanning . Vienna, Austria: International Atomic Energy Agency, 1959:59.
[3] Birks JB. The theory and practice of scintillation counting. New York: MacMillan, 1964:432.
[4] United States Pharmacopeia (USP). 1999:91-93. Niemeyer MG, Kuijper AF, Gerhards LJ et al. Nitrogen-13 ammonia perfusion Imaging: relation to meta imaging. AM Heart J 1993; 125:848-854.
[5] Ak I, stokkel MPM, Pauwels EKJ. Position emission tomography with 2-[18F] flouoro-2-deoxy-D-glucose in oncology. J Cancer Res Clin Oncol 2000;127:2-8.
[6] B. ML. The role of radionuclide in the patient with osteogenic sarcoma. Semin Roentgenol 1989; 24:185.
[7] R. EL, Schellhanner F. Solitary lesions on bone scan in genitourinary malignancy. J Urol 1984; 132:1000-1002.
[8] L. Garvican and S. Field, “A pilot evaluation of the R2 image checkersystem and users’ response in the detection of interval breast cancers on previous screening films,” Clin. Radiol., vol. 56, no. 10, pp. 833—837, Oct. 2001.
[9] I. Leichter, S. Fields, R. Nirel, P. Bamberger, B. Novak, R. Lederman,and S. Buchbinder, “Improved mammographic interpretation of massesusing computer-aided diagnosis, ” Eur. Radiol., vol. 10, no. 2, pp. 377—383, 2000.
[10] B. Verma and J. Zakos, “A computer-aided diagnosis system for digital mammograms based on fuzzy-neural and feature extraction techniques,” IEEE Trans. Inform. Tech. Biomed., vol. 5, pp. 46—54, Mar. 2001.
[11] H. D. Cheng,Y. M. Lui, and R. I. Freimanis, “Anovel approach to microcalcification detection using fuzzy logic technique,” IEEE Trans. Med.Imag., vol. 17, pp. 442—450, June 1998.
[12] S. Yu and L. Guan, “A CAD system for the automatic detection of clustered microcalcification in digitized mammogram films,” IEEE Trans. Med. Imag., vol. 19, pp. 115—126, Feb. 2000.
[13] H. Li, Y. Wang, K. J. R. Liu, S. C. B. Lo, and M. T. Freedman, “Computerized radiographic mass detection-Part I: Lesion site selection by morphological enhancement and contextual segmentation,” IEEE Trans. Med. Imag., vol. 20, pp. 289—301, Apr. 2001.
[14] H. Li, Y. Wang, K. J. R. Liu, S. C. B. Lo, and M. T. Freedman, “Computerized radiographic mass detection-Part II: Decision support by featured database visualization and modular neural networks,” IEEE Trans. Med. Imag., vol. 20, pp. 302—313, Apr. 2001.
[15] M. G. Penedo, M. J. Carreira, A. Mosquera, and D. Cabello, “Computer-aided diagnosis: A neural-network-based approach to lung nodule detection,” IEEE Trans. Med. Imag., vol. 17, pp. 872—880, Dec. 1998.
[16] K. Sato, K. Sugawara, Y. Narita, and I. Namura, “Consideration of the method of image diagnosis with respect to frontal lobe atrophy,” IEEE Trans. Nucl. Sci., vol. 43, pp. 3230—3239, Dec. 1996.
[17] Y. E. Erdi, John L. Human, Massimo Imbriaco, H. Yeung and S. M. Larson. Quantitative Bone Metastases Analysis Based on Image Segmentation. J Nucl Med 1997; 38:1401-1406.
[18] T.-K. Yin, N.-T. Chiu, A Computer-Aided Diagnosis for Locating Abnormalities in Bone Scintigraphy by a Fuzzy System With a Three-Step Minimization Approach. IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 23, NO. 5, MAY 2004
[19] O. J. Tobias, IEEE, and R. Seara, Image Segmentation by Histogram Thresholding Using Fuzzy Sets. IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 11, NO. 12, DECEMBER 2002
[20] M. CE. Basic principles of ROC analysis. Semin Nucl Med 1988; 8:283-298.
[21] M. CE. ROC m

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔