Showing papers by "Guanyu Yang published in 2006"
TL;DR: A new fast method for computing the Legendre moments is proposed, which is not only suitable for binary images but also for grey level images, and it is shown that the proposed method is more efficient than the direct method.
49 citations
01 Jan 2006
TL;DR: The proposed approach makes use of a tracking algorithm of the vascular structure, combining a 3D geometric moment operator with a multiscale Hessian filter to estimate the vessel central axis location, its local diameter and orientation.
Abstract: This paper deals with the extraction of the coronary network on dynamic volume sequences, acquired in Multi-slice Spiral Computed Tomography (MSCT). The proposed approach makes use of a tracking algorithm of the vascular structure, combining a 3D geometric moment operator with a multiscale Hessian filter to estimate the vessel central axis location, its local diameter and orientation. The method performs at the same time, a bifurcation detection to reconstitute the structure of the coronary network. The mean computation time to extract a coronary network is about 3 minutes using a P4-2.4G PC. Preliminary encouraging results are presented on one volume of a sequence.
29 citations
TL;DR: A method for the coronary artery extraction and the lesion quantification makes use of a two stage-process that was submitted to a clinical evaluation on six data sets that showed good sensitivity and specificity performances.
Abstract: The last technological advances of multi-detector computed tomography (MDCT) have brought significant improvements in image quality and consequently diagnostic performance. It opens, in particular, new perspectives for the coronary artery disease assessment. This paper presents a method for the coronary artery extraction and the lesion quantification. It makes use of a two stage-process. A tracking algorithm of the vascular structure, based on a 3D geometric moment operator, is first applied to estimate the vessel central axis location and its local diameter. It led to the extraction of the overall coronary tree. A deformable model technique, using a modified version of 3D level set technique, is then run on pathological vascular segments to refine the delineation of the wall and parietal calcifications. Parameters such as the narrowing or enlargement ratio of the vascular lumen, the minimum/maximum areas and the percent-area stenosis or aneurysm, are then computed for diagnosis purpose. The first method was submitted to a clinical evaluation on six data sets that showed good sensitivity and specificity performances. The second one was validated by experts, through a visual inspection of 2D cross-sections and volume rendering.
8 citations
01 Jan 2006
TL;DR: An algorithm is proposed that perform a temporal tracking of the vessel central axis in a 3-D dynamic sequence in multi-slice computed tomography (MSCT) based on geometric moments and a local cylindrical approximation.
Abstract: An algorithm is proposed that perform a temporal tracking of the vessel central axis in a 3-D dynamic sequence in multi-slice computed tomography (MSCT). The approach is based on geometric moments and a local cylindrical approximation. The local characteristics of the vessel are estimated on the first volume of the sequence (position on the vessel central axis, local diameter, intravascular and background intensities), then used to track the vessel along the sequence. The correspondence between two volumes is solved through a region matching based on a criterion of minimal distance combining moment-based descriptors with intensity information. Preliminary results are presented on two sequences.
6 citations