¡¡Chinese Journal of Computers   Full Text
  Title3-D Reconstruction of Coronary Arterial Trees from Uncalibrated Monoplane Angiographic Images
  AuthorsYANG Jian1) WANG Yong-Tian1) LIU Yue1) TANG Song-Yuan1) LI Yan-Hui2)
  Address1)(Beijing Institute of Technology, Beijing 100081)
2)(Heart Center, Beijing Red-cross Chaoyang Hospital Affiliated to Capital University of Medical Sciences, Beijing 100020)
  Year2007
  IssueNo.9(1594¡ª1602)
  Abstract &
  Background 		Abstract This paper presents a novel method for the 3-D reconstruction of coronary arterial trees from two uncalibrated monoplane angiographic images. This method does not require independent calibration of the phantom. The lower-order distortion, transformations of the C-arm and the movement of the table can be effectively estimated. This method is very flexible because it can not only be applied to images with the same focal lengths, but also can be used to images with different focal lengths. Experimental results show that the proposed method is very robust and effective for the reconstruction of coronary arterial trees, and is able to achieve less than 0.26mm of back-projection RMS errors, which represents a great improvement on the accuracy of the 3-D reconstruction of coronary arteries from uncalibrated images.

keywords coronary artery; 3-D reconstruction; monoplane angiographic image; calibration; distortion correction

background The research is part of the project "DSA Image Intelligent Analysis and 3-D Reconstruction of Coronary Arteries" supported by the National Basic Research Program (973 Program) of China under grant No.2003CB716105. 3-D reconstruction of vessels from digital X-ray angiographic images is a powerful technique to make up for the limitation of the instrument of angiography. It can provide doctors with accurate inspection of the complex arterial network and quantitative assessment of vascular diseases in three dimensions. Based on in-depth studies of the imaging principle of the X-ray angiographic instrument and theories of computer vision, computer graphics and digital image analysis, using real coronary X-ray angiographic images and limited imaging parameters, this work aims to develop effective methods for X-ray angiographic image analysis and 3-D reconstruction of coronary arteries.