Abstract | ||
---|---|---|
Coronary artery lesions are highly correlated with their movement patterns. However, until now few reports have been devoted to coronary motion analysis based on Computed Tomography Angiography (CTA). In this paper, a registration method is proposed to match the 3D coronary point sets between adjacent cardiac phases, derived from CTA. The proposed method considers the alignment problem as the estimation of a mixture probability density and makes full use of the global and local structure information of the point sets. Furthermore, the landmark points in some specific positions, such as coronary artery bifurcations, are exploited as a priori knowledge to impr-ove the matching accuracy. The movement function and corres-ponding relationship between two coronary point sets are obtained by the Expected Maximum (EM) algorithm. Exper-imental results demonstrate the effectiveness of the proposed algorithm for deformed and missing coronary point sets.
|
Year | DOI | Venue |
---|---|---|
2019 | 10.1145/3364836.3364875 | Proceedings of the Third International Symposium on Image Computing and Digital Medicine |
Keywords | Field | DocType |
CTA, Coronary artery, Expected Maximum, Global/ Local structure information, Landmark, Point sets, Registration | Computer vision,Point set registration,Computed tomography angiography,Computer science,A priori and a posteriori,Local structure,Artificial intelligence,Motion analysis,Landmark,Probability density function,Match moving | Conference |
ISBN | Citations | PageRank |
978-1-4503-7262-6 | 0 | 0.34 |
References | Authors | |
0 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Bu Xu | 1 | 0 | 0.34 |
Lisheng Xu | 2 | 178 | 39.09 |
Ying Kang | 3 | 0 | 0.34 |
Along Song | 4 | 0 | 0.68 |
Bin Wang | 5 | 0 | 0.68 |
Lu Wang | 6 | 3 | 1.57 |
Benqiang Yang | 7 | 0 | 0.34 |