- 學位論文
心導管影像之三維重建與計算流體力學之應用
Three Dimensional Lumen Reconstruction and Computational Fluid Dynamics Application of Coronary Angiography
摘要
冠狀動脈疾病是最常見的心血管疾病，並且為全球最致命的疾病。而心導管檢查是最有效的冠狀動脈疾病檢查手段，冠狀動脈造影 (coronary angiography image, CAG)影像以及血流儲備分數(fractional flow reserve, FFR)會在檢查中取得。本研究應用影像三維重建演算法，利用多個冠狀動脈造影影像重建出三維冠狀動脈結構。並與電腦斷層冠狀動脈血管攝影(coronary computed tomography angiography, CCTA)中的血管結構做對位，結合兩影像診斷工具之優勢，電腦斷層冠狀動脈血管攝影可以提供斑塊的資訊，而冠狀動脈造影擁有更準確的管徑資訊，兩影像融合的結果可以提供診斷上更多的參考依據。計算流體力學(Computational fluid dynamics, CFD)方法將用於模擬動脈內的壓力分佈，作為判斷心肌缺血可能性的依據。兩種不同的血液質量流率估算方法將於本研究中應用，模擬出的結果可作為判斷冠狀動脈疾病之依據。其中方法一從冠狀動脈造影影像中提取出血液流速資訊，因不需要電腦斷層冠狀動脈血管攝影資訊，使其應用上的限制更少，而方法二從電腦斷層冠狀動脈血管攝影影像中提取出心肌大小資訊，擁有更好的分類準確率。透過上的述重建演算法，所得到的結果可應用於多個面相，藉以輔助冠狀動脈疾病檢查。
並列摘要
Coronary artery disease (CAD) is the most common type of heart disease and is the leading cause of death globally. Cardiac catheterization is a helpful invasive procedure for CAD diagnosis and treatment. The coronary angiography (CAG) image and fractional flow reserve (FFR) values are acquired during cardiac catheterization. The study applied a three-dimensional (3D) reconstruction algorithm to generate the coronary artery geometry from CAG images. The registration of 3D CAG and CCTA artery geometry is applied to combine the advantages of these two image modalities. Since the plaque is visible in CCTA, and CAG provides accurate lumen radius information, the fusion results can proffer more diagnosis information. Computational fluid dynamics (CFD) was used to simulate the distribution of coronary arterial pressure and evaluate the likelihood of myocardial ischemia. Two methods were used to estimate the mass flow rate in this study. The simulation results can be used to classify the severity of CAD. Method 1 extracted the blood velocity information from CAG images, and Method 2 extracted the size of the cardiovascular system from CCTA images. Method 2 has better classification accuracy. CCTA is unnecessary for Method 1, which makes it more flexible. Through the reconstruction algorithm described above, the results can be applied in various ways to help the CAD diagnosis.