Muhit Abdullah A, Pickering Mark R, Ward Tom, Scarvell Jennie M, Smith Paul N
School of Engineering and Information Technology, The University of New South Wales at the Australian Defence Force Academy, Canberra, Australia.
Annu Int Conf IEEE Eng Med Biol Soc. 2010;2010:6288-91. doi: 10.1109/IEMBS.2010.5628083.
3D computed tomography (CT) to single-plane 2D fluoroscopy registration is an emerging technology for many clinical applications such as kinematic analysis of human joints and image-guided surgery. However, previous registration approaches have suffered from the inaccuracy of determining precise motion parameters for out-of-plane movements. In this paper we compare kinematic measurements obtained by a new 2D-3D registration algorithm with measurements provided by the gold standard Roentgen Stereo Analysis (RSA). In particular, we are interested in the out-of-plane translation and rotations which are difficult to measure precisely using a single plane approach. Our experimental results show that the standard deviation of the error for out-of-plane translation is 0.42 mm which compares favourably to RSA. It is also evident that our approach produces very similar flexion/extension, abduction/adduction and external knee rotation angles when compared to RSA.
从三维计算机断层扫描（CT）到单平面二维荧光透视配准是一项新兴技术，可用于许多临床应用，如人体关节的运动学分析和图像引导手术。然而，以前的配准方法在确定平面外运动的精确运动参数时存在不准确的问题。在本文中，我们将一种新的二维-三维配准算法获得的运动学测量结果与金标准伦琴立体分析（RSA）提供的测量结果进行了比较。特别是，我们对平面外平移和旋转感兴趣，而使用单平面方法很难精确测量这些参数。我们的实验结果表明，平面外平移误差的标准差为0.42毫米，与RSA相比具有优势。同样明显的是，与RSA相比，我们的方法产生的屈伸、外展/内收和膝关节外旋角度非常相似。
