基於有標記點之立體影像以有限元素法模型模擬肝臟形變與切割

Abstract

在微創手術中，醫師只能以內視鏡提供的平面影像作為導引，配合手術前的電腦斷層掃瞄影像自行在腦中推斷肝臟中血管與腫瘤的位置分布情形，但肝臟在手術過程中會因醫師的翻動動作而產生形變，甚至是切割使肝臟外型產生變化，都會使對肝臟內部血管與腫瘤的位置的掌握更加困難。藉由模型模擬計算血管位置，醫師可以更直觀的掌握內部血管與腫瘤的分布，降低切到大血管或腫瘤的風險。 在本論文中，利用電腦斷層掃瞄影像建立具有生物力學特性的有限元素法模型模擬手術中肝臟因為變形與切割而造成的外型變化，藉由在肝臟表面放置標記點，並利用具深度資訊的內視鏡影像獲得手術中肝臟部分表面的影像，分析手術中(變形與切割後)及手術前(未變化)的肝臟表面標記點位移，同時利用演算法分析表面幾何特徵點差異來模擬肝臟外型變化，在有限元素法模擬計算後可以得到變形與切割後肝臟內部血管與腫瘤分布情形。我們利用模擬資料及體外豬肝實驗驗證形變計算並利用模擬資料驗證切割計算。

In minimally invasive surgery (MIS), surgeon reviews the CT scan images before surgery to know the position of vessels and tumors. During the surgery, the video stream from endoscopic camera is the only information of liver structure providing to surgeon. However, the preoperative information is not trustworthy because liver is deformed or even cut by different surgical procedures. In this thesis, a finite element model is used to calculate the deformation and cutting of live during the surgery. Preoperative model is built from CT image, and intraoperative data is partial liver surface from RGB-D endoscopic camera. We use marker spread on the liver surface to register intraoperative data to preoperative model and compute the model structure change to simulate the surgical procedure. After finite element computation, the intraoperative vessels and tumor position is acquired. With the help of this internal structure computation system, surgeon can operate surgical instruments more efficiency and reduce the risk. Our approach is validated through simulation and ex vivo experiment for performance of deformation and simulation for performance of cutting.