臺灣博碩士論文加值系統

肝臟腫瘤分列為男性第一位及女性第二位的高致死率。傳統上使用理學檢查、實驗室血液生化檢查以及影像學方面包括電腦斷層、超音波、磁振造影等方式的檢查做為診斷的依據。本研究藉由擴散加權影像造影(diffusion weighted imaging; DWI)在超順性氧化鐵離子造影劑(super paramagnetic iron oxide; SPIO)注射後，並在有適當的影像後處理程序下，提高評估肝臟腫瘤上的正確性。
本研究篩選疑似肝臟腫瘤病患10人做為本研究取樣的對象，實施傳統肝臟磁振造影，以及SPIO注射前所實施的DWI肝臟磁振造影，之後實施一組與DWI波序在幾何條件上完全相符的T1W影像，以提供解剖資訊上影像融合的基礎，並在SPIO注射後10分鐘、20分鐘及30分鐘後實施DWI肝臟磁振造影。為評估DWI肝臟磁振造影的特異性(specificity) ，我們亦選取五名無肝膽系統疾病之健康志願者，實施傳統肝臟磁振攝影及SPIO注射前的DWI肝臟磁振造影，做為實驗的對照組。再以mathlab影像處理程式，運用影像分析的方式，將所獲得的肝臟擴散加權影像，分割(segmentation)並以適當的數位濾波(digital filter)強化影像的邊緣（edge enhancement），以繪製出病灶的範圍，並T1加權影像Fusion 在一起，藉此同時獲得功能性與解剖學上的資訊。此外，兩名具專科醫師資格的放射線診斷醫師個別獨立的對這些傳統磁振攝影影像進行判讀。
初步結果顯示，本研究所採取的技術，在所有符合評估標準的八名受測病患中皆能準確的偵測出病灶的位置，其敏感度(sensitivity)為1；同時在五名正常志願者的測試中，DWI肝臟磁振造影亦無偵測出假陽性病灶，其特異性(specificity)為1。此外，各判讀醫師對此技術的滿意度亦高達93.75%。
我們預期此一結合SPIO-enhanced MR imaging與DWI的技術，作為一種輔助性的工具，應可普遍應用在臨床MRI造影上。配合傳統肝臟磁振造影及SPIO-enhanced MR imaging的實施相信對臨床上肝臟病灶的偵測上應有絕對的幫助，同時亦可大幅提高臨床上肝臟病灶的偵測率。

Hepatic tumor was the top of deadly diseases for male and the second deadly diseases for female in Taiwan. Traditional, physical examination, laboratory blood test and medical imaging including computer tomography (CT), ultrasound and magnetic resonance imaging (MRI) were used for hepatic lesions diagnosis. The purpose of this study is to evaluate the accuracy of hepatic lesion detection after Super paramagnetic iron oxide (SPIO) injection followed by diffusion weighted imaging (DWI).
Ten patients with suspected hepatic tumors were recruited for this study. For each patient, conventional MR liver scanning and pre-SPIO diffusion weighted imaging (DWI) were performed before contrast agent administration. Additional, a T1 weight (T1W) imaging series with the same geometric characterizes was performed for images fusion. Then 10min, 20min, 30min after SPIO administration DWI were performed. Five volunteers without any biliary system diseases were chosen to perform conventional MR liver scanning and pre-SPIO diffusion weighted imaging (DWI) to evaluate the specificity in this study. Moreover, an image processing program was used to segmentation and enhances the edge of hepatic tumor; T1W images were fused with these post-processed DWI to archive both anatomic and functional information. Then, these images were elevated independently by two radiologists with blind patient’s clinical information
The primary result shows that all the present lesions of the eight patients could be detected, i.e. the sensitivity is 1; and during the volunteer group test, no any false positive lesion has been detected, therefore the specificity is 1. This technique also has been reviewed by radiologists and the satisfaction level is up to 93.75%.
Therefore, to be an assisted tool, we expect that by combining this technique with SPIO-enhanced MR imaging and DWI, can become a popular method for hepatic lesions detection in MR scanning. Coordinated with SPIO-enhanced MR imaging and DWI, SPIO-enhanced MR imaging, we believe this technique is useful and can increase the ability of lesion detection ratio significantly.

中文摘要--------------------------------------------------------------------------------- I
英文摘要--------------------------------------------------------------------------------- III
誌 謝---------------------------------------------------------------------------------- V
目 錄-------------------------------------------------------------------------------- VII
圖 索 引----------------------------------------------------------------------------------IX
表 索 引---------------------------------------------------------------------------------- X
縮 寫 表--------------------------------------------------------------------------------- XI

第一章 緒論------------------------------------------------------------------------------ 1
1-1 前言----------------------------------------------------------------------------------- 1
1-2 研究背景----------------------------------------------------------------------------- 1
1-3 文獻回顧----------------------------------------------------------------------------- 3
1-4 研究目的----------------------------------------------------------------------------- 4
1-5 論文架構----------------------------------------------------------------------------- 6

第二章 理論基礎------------------------------------------------------------------------ 7
2-1 肝臟的解剖以及其功能----------------------------------------------------------- 7
2-1-1 肝癌的可能發病原因----------------------------------------------------------- 8
2-1-2 臨床上的診斷-------------------------------------------------------------------- 9
2-2 磁振造影的基本成像原理------------------------------------------------------ 11
2-3 敏感度編碼技術------------------------------------------------------------------ 13
2-3-1 敏感度編碼技術在擴散加權影像中的應用------------------------------- 13
2-4 擴散加權影像--------------------------------------------------------------------- 14
2-5 擴散相關係數 ( b value)-------------------------------------------------------- 16
2-6 邊緣偵測的基本原理------------------------------------------------------------ 17

第三章 研究架構與方法------------------------------------------------------------- 19
3-1 取樣的來源------------------------------------------------------------------------- 19
3-1-1 研究對象的選擇---------------------------------------------------------------- 19
3-1-2 實驗的對照組------------------------------------------------------------------- 19
3-2 研究流程--------------------------------------------------------------------------- 20
3-2-1 傳統磁振攝影------------------------------------------------------------------- 21
3-2-2 擴散加權影像------------------------------------------------------------------- 23
3-2-3 實施的步驟---------------------------------------------------------------------- 23
3-3 影像的分析------------------------------------------------------------------------ 24
3-3-1 磁振影像的擷取--------------------------------------------------------------- 24
3-3-2 高通濾波器的選擇------------------------------------------------------------ 25
3-3-3 擴散加權影像的呈現--------------------------------------------------------- 26
3-4 影像的評估------------------------------------------------------------------------ 27

第四章 結果與討論------------------------------------------------------------------- 29
4-1 研究結果--------------------------------------------------------------------------- 29
4-1-1 初步實驗成效的驗證---------------------------------------------------------- 29
4-1-2 實驗前期的結果---------------------------------------------------------------- 34
4-1-3 實驗參數的改善---------------------------------------------------------------- 35
4-1-4 最終實驗的結果---------------------------------------------------------------- 36
4-1-5 SPIO 注射前後擴散加權影像的比較--------------------------------------- 43
4-2 討論---------------------------------------------------------------------------------- 44
4-2-1 一般概論------------------------------------------------------------------------- 44
4-2-2 EPI 波序的優勢----------------------------------------------------------------- 45
4-2-3 影像融合中的假影------------------------------------------------------------- 46
4-2-4 b 值對影像的影響-------------------------------------------------------------- 47
4-2-5 不同時間點取得的DWI影像------------------------------------------------49
4-2-6 臨床上的限制---------------------------------------- ---------------------------50
4-2-7 病患本身的限制---------------------------------------------------------------- 51
4-2-8 影像融合後所帶來的效益---------------------------------------------------- 53

第五章 結論與未來展望------------------------------------------------------------- 55
5-1 結論--------------------------------------------------------------------------------- 55
5-2 未來展望--------------------------------------------------------------------------- 56

參考文獻-------------------------------------------------------------------------------- 57

附錄-------------------------------------------------------------------------------------- 59
附錄一、 醫師評分判讀說明------------------------------------------------------ 59
附錄二、肝臟磁振檢查病灶偵測評分量表-------------------------------------- 62
附錄三、SPIO注射後並經後影像處理之擴散加權影像成效評估表------- 63
附錄四、不同組別影像的量化分數分析長條圖-------------------------------- 64
圖索引
圖1. 人體肝臟的正面及背面圖----------------------------------------------------- 7
圖2. 不同組織的T1(a)及T2 (b)時間--------------------------------------------- 12
圖3. EPI波序示意突及快速轉換的梯度磁場 ---------------------------------- 15
圖4. 擴散加權影像的示意圖------------------------------------------------------- 15
圖5. 研究的流程圖------------------------------------------------------------------- 20
圖6. 1.5T AVANTO MR Scanner 及腹部陣列線圈----------------------------- 21
圖7.SPIO 注射前後波序實施與時間的示意圖--------------------------------- 24
圖8.幾何條件上完全相符的DWI影像以及T1W影像----------------------- 25
圖9.以不同的濾波器所做出的邊緣強化影像----------------------------------- 25
圖10.相同病患以不同方式所呈現的後處理影像-------------------------------26
圖11.SPIO-enhanced 前所實施的磁振掃描影像--------------------------------30
圖12.SPIO-enhanced後所實施之T1 vibe 3D 動態影像-----------------------31
圖13.T2 haste with short TI recovery transverse section的影像-----------------------------------------------------------------------------------31
圖14.Diffusion weighted imaging的影像------------------------------------------32
圖15.SPIO注射後之肝臟DWI經後處理軟體分析之結果--------------------33
圖16.DWI影像、T1影像 以及Fusion 後的影像------------------------------34
圖17.SPIO注射後並經後影像處理之擴散加權影像之滿意度調查結果-----------------------------------------------------------------------------------42
圖18.SPIO注射前後擴散加權影像比較的長條圖------------------------------ 44
圖19.數位濾波濾除雜訊前後的影像-----------------------------------------------46
圖20.不同b值在訊號上的影響-----------------------------------------------------48
圖21.不同b值在肝臟磁振造影中對訊號的影響--------------------------------48
圖22.水訊號太強導致病灶訊號的對比較低的影像-----------------------------52
表索引
表1. 病患取樣的基本資料---------------------------------------------------------- 38
表2. 傳統磁振攝影影像的量化結果---------------------------------------------- 39
表3. SPIO 注射前的DWI SENSE imaging影像的量化結果----------------- 39
表4. SPIO注射後的 DWI SENSE imaging影像的量化結果----------------- 40
表5. SPIO注射後並經後影像處理的DWI SENSE imaging影像的量化結果------------------------------------------------------------------------------------- 41
表6. 依據『SPIO注射後並經後影像處理之擴散加權影像評估表』 的評估結果---------------------------------------------------------------------------- 42
表7. SPIO注射前後擴散加權影像的比較----------------------------------------43
表8. 不同b值在肝臟磁振造影中對訊號的影響--------------------------------59

1. Nasu K, Kuroki Y, and Nawano S: Hepatic metastases: Diffusion-weighted Sensitivity-encoding versus SPIO-enhanced MR Imaging; Radiology 2006; 239:122-130.
2. Okano K, Maeba T, Ishimura K, et al: Hepatic resection for metastatic tumors from gastric cance; Ann Surg 2002; 235:86–91.
3. Gore JC and Majumdar H: Super paramagnetic Iron Oxide Particles, Radiology 1988; 169:657-658
4. Kim MJ, Focal Hepatic lesion: Detection and characterization with combination Gadolinium and SPIO enhanced MR imaging, Radiology 2003; 228:719-726.
5. Warach S, Chien D, and Li, W: Fast magnetic resonance diffusion-weighted imaging of acute human stroke, Neurology 1992; 42(9):1717-1723
6. Chiu FY, Jao JC, Chen CY, et al. Effect of intravenous Gadolinium-DTPA on diffusion-weighted magnetic resonance images for evaluation of focal Hepatic lesions; Journal of Computer Assist Tomography 2005; 29:176-180.
7. Deng JR, Thomas K, and Sato KT: In vivo diffusion-weighted imaging of liver tumor necrosis in the VX2 rabbit model at 1.5 Tesla; Investigative Radiology April 2006; 41(4):410-414.
8. Kon K Y, Kwak, K H, and Han Y M, Usefulness of Combining Sequentially Acquired Gadobenate Dimeglumine-Enhanced Magnetic Resonance Imaging and Resovist-Enhanced Magnetic Resonance Imaging for the Detection of Hepatocellular Carcinoma: Comparison With Computed Tomography Hepatic Arteriography and Computed Tomography Arterioportography Using 16-Slice Multidetector Computed Tomography; Journal of Computer Assist Tomography 2007; 31(5):702-711.
9. Madore B, Pelc NJ. SMASH and SENSE: experimental and numerical comparisons. Magn Reson Med. 2001; 45:1103-1111.
10. 財團法人肝病防治學術基金會, 全球肝病資訊網 2008/10/1, http://liver.org.tw/
11. Ray H. Hashemi, Chapter 3: Radio Frequency Plus, MRI the Basics, p32-40, 2004, Williams & Wilkins. U.S.A.
12. Ray H. Hashemi, Chapter 4: T1,T2 and T2*, MRI the Basics, p41-48, 2004, Williams & Wilkins. U.S.A.
13. Ray H. Hashemi, Chapter 22: Echo Planar Imaging, MRI the Basics, p263-273, 2004, Williams & Wilkins. U.S.A.
14. Alasdair McAndrew,（徐曉珮 譯） Introduction to Digital Image Processing with MATLAB（數位影像處理）:（高立圖書有限公司），2005:255-272
15. John F. Canny. A computational approach to edge detection; IEEE Transactions on Pattern Analysis and Machine Intelligence 1986; 8(6):679-698
16. Patrick A, Patrick A H, Alto S, Free-Breathing Echo-Planar Imaging BasedDiffusion-Weighted Magnetic Resonance Imaging of the Liver With Prospective Acquisition Correction. J Comput Assist Tomogr .2008; 32:372-378
17. Donald W.McRobbie, Elizabeth A. Moore, Martin J. Graves et al. Chapter 5: Basic image optimization, MRI from Picture to Proton, Second edition, P65-78, 2007, Cambridge.