醫療影像重建與手術模擬系統之研究; 3D Reconstruction via Medical Images and Laparoscopic Surgery Simulation System

NCU Institutional Repository > 工學院 > 機械工程研究所 > 博碩士論文 > Item 987654321/1826

請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/1826

題名:	醫療影像重建與手術模擬系統之研究;3D Reconstruction via Medical Images and Laparoscopic Surgery Simulation System
作者:	宋文旭;Wen-Hsu Sung
貢獻者:	機械工程研究所
關鍵詞:	虛擬實境;模型重建;體積計算;手術模擬;系統驗證;3-D reconstruction;volume estimation;system verification;laparoscopic surgery simulation;virtual reality
日期:	2003-01-18
上傳時間:	2009-09-21 11:32:34 (UTC+8)
出版者:	國立中央大學圖書館
摘要:	本研究結合虛擬實境技術，發展一可由電腦斷層或核磁共振等醫療影像，重建選取區域之3D模型，並提供高準確度體積計算之系統。完成開發程序後並設計相關之驗證實驗，將數個已知外觀與體積之塑膠人體器官模型，實際進行核磁共振造影，以進行各模型之外觀重建與體積計算測試。由測試結果可以發現，各模型之3D外觀重建結果與真實外觀非常近似，而體積計算結果亦與各模型以排水法量測所得之體積數值相當接近。完成體積計算結果準確性之實驗驗證後，接著與國內某醫學中心合作，以三種不同的體積計算方式，包含單一最大徑、橢圓法與本研究之方法，進行腫瘤體積與各項子宮頸癌預後因子之相關性實驗。由實驗結果可以發現，以本研究之方式計算腫瘤體積，較其他兩種傳統的方法，可以提供更準確的子宮頸癌的骨盆淋巴節轉移與否、是否有子宮旁側軟組織侵犯與腫瘤細胞分化狀況之初步預估。本研究並以虛擬實境技術，開發設計一腹腔鏡手術模擬訓練系統，可進行雙極電燒輸卵管結紮手術之全程模擬訓練。在完成系統之開發工作後，並規劃了學習性、信度與不同訊息回饋對使用者學習成效影響之相關驗證實驗。由學習性實驗結果可以得知，模擬系統具備可學習性，受試者在經過8次的練習後，其操作行為將可達到一熟練穩定的狀態。由信度實驗結果可以得知，讓同一批受試者在經過學習性實驗後約7至14天，再進行相同的測試，實驗結果數據與學習性實驗相當接近，由此可證明本系統具備信度，不會因操作時間的改變而導致不同的結果。而由不同訊息回饋對使用者學習成效影響之實驗可以發現，以文字、音效與輔助側視圖三種不同的訊息回饋方式，對使用者之平均操作時間、平均錯誤次數與平均夾合次數皆可產生顯著的影響。最後則將已開發之腹腔鏡手術模擬訓練系統進行功能擴充，整合具備力回饋功能之模擬手術刀具，並設計腹腔鏡頭之模擬操控方式，以期可以提供使用者更接近真實腹腔鏡手術過程之模擬操作情境。完成系統功能之擴充改善工作後，隨後並再以學習性、信度與不同訊息回饋實驗進行使用者之操作行為與學習成效之驗證實驗。由學習性與信度實驗結果可以發現，經功能擴充後的模擬系統具備學習性與可信度；然而，模擬系統新增之力回饋與腹腔鏡頭模擬操控功能，因訊息功能本身設計與實驗規劃之緣故，對使用者之學習成效指標，包含平均操作時間、平均錯誤次數與平均夾合次數等等，並無產生顯著之影響。 The study integrated virtual reality technology and developed a system which would reconstruct three-dimensional (3-D) models and provide highly accurate volume estimations from selected areas of medical images. After development processes, we devised an experiment to test and verify the results of 3-D reconstruction and volume estimations. The experiment result showed that the appearances of reconstruction models were very similar to the real ones, and the value of volume estimations were very close to the values evaluated by drainage. After verifying the accuracy of volume estimations, another clinical experiment was implemented. Three kinds of tumor volume estimations, including the single maximum diameter method, ellipse method and our 3-D integration method, were performed and the relationship of these volume estimations among associated cervical cancer prognostic parameters was evaluated.This study found that the parameters of lymph nodes metastasis, parametrial involvement and tumor differentiation are volume dependent. 3D-tumor volumetry showed superior discrimination of these parameters than current single maximum diameter evaluation suggesting its potential as a rapid method for initial prediction of prognostic factors in cervical cancer. In this study, a virtual reality-based simulator system was developed for extensive laparoscopic surgery training. The purpose of this study is to assess the feasibility of virtual reality-based laparoscopic gynecology simulation system. Ten healthy, non-disabled volunteers were recruited. The surgical procedure is a process of tubal sterilization by cauterization. Volunteers followed the training procedure fifteen trials in the first test and retest respectively. Stable performances were obtained after about eight trials for all subjects. The results of this study indicate that the system is stable and the system has fair high test-retest reliability. Another thirty two non-disabled volunteers were recruited for participating in the different signal feedback experiment. The purpose of this experiment is to assess the influence of text, sound and lateral view cues on the manipulation performance in the system. The average of task time, error times and clip times are the parameters used to evaluate the manipulation performance. It can be concluded that all of the text, sound and lateral view cues are significant cues of manipulation performance of this simulation system. Finally, we employed another laparoscopic gynecology simulating device with force feedback functions, and devised a new function of changing the field of view of laparoscope camera for enhancing the capability of this simulation system. The feasibility and different signal feedback experiments were executed again. The results of the experimants indicate that the new system is stable and has fair high test-retest reliability. Because of the effects of the functions and design of this experiment, the experiment results shows that the new functions, including both force feedback and changing the field of view of laparoscope camera, are not significant cues of manipulation performance of this simulation system.
顯示於類別:	[機械工程研究所] 博碩士論文

文件中的檔案:

檔案	大小	格式	瀏覽次數

在NCUIR中所有的資料項目都受到原著作權保護.

社群 sharing

資料載入中.....