In-vivo measurement of Stress-Strain Distribution on the Humnan Ligament using Arthroscopic Images

本研究旨在建立人体膝关节韧带表面应变分布的在体关节镜测量系统。为了解决活体的技术和伦理问题，我们计划使用微创关节镜的单平面图像进行分析。我们应用了改进的运动捕捉系统，通过该系统，我们将能够通过关节镜检查检测标记（ACL表面的骨骼或纹理）的3D位置。然而，当我们应用该技术测量ACL应变时，我们发现可检测到。因此，作为一种替代方法，我们应用了阴影形状方法。使用水平集方法，使用处理后图像的亮度数据计算三维坐标数据。并利用B样条曲线再现韧带表面的三维形状。计算所需的所有参数数据均从校准图像中引入。我们的方法的准确性进行了检查，从已知的主题。从而产生足够的准确性。

The object of this study is to establish in vivo arthroscopy measurement system of surface strain distribution of the human knee ligaments. To solve technical and ethical problems subject to living body, we planned to use single plane image from less invasive arthroscopy for the analysis. We applied the modified motion capture system by which we would be able to detect the 3D positions of markers (skecles or textures on the ACL's surface) through an arthroscopic examination. When we applied the technique to measure the ACL strain however, we found the detectable Thus as an altenative, we applied the Shape from Shading method. Three-dimensional coordinate data were calculated using Level Set Method, using the brightness data of the processed image. And three-dimensional shape of ligament surface was reproduced by B-Spline. All parametric data necessary for calculation were introduced from the calibration image. The accuracy of our method was checked from known subjects. and thereby resulting in enough accuracy.