Name
Title | ||
|---|---|---|
Automated Cardiovascular Segmentation In Patients With Congenital Heart Disease From 3d Cmr Scans: Combining Multi-Atlases And Level-Sets |
Abstract | ||
|---|---|---|
This paper presents an automatic method that enables segmentation of the whole heart and the great vessels from 3D MRI scans. The proposed method is built upon a multi-atlas-based segmentation approach and consists of a number of intermediate steps to enable accurate segmentation of the ventricular myocardial tissue, intra-cardiac blood pool, and the aorta. The method was tested on the datasets provided by the HVSMR 2016 MICCAI challenge organizers. Results on the testing data show that, the proposed method achieved an average Dice index of 0.89 for the blood pool and 0.75 for the ventricular myocardial tissue. The corresponding average surface distance error were 1.6mm and 1.1 mm, respectively. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.1007/978-3-319-52280-7_15 | RECONSTRUCTION, SEGMENTATION, AND ANALYSIS OF MEDICAL IMAGES |
Keywords | DocType | Volume |
Atlas-based segmentation, Registration, Tissue maps, Hough transform, Level-sets | Conference | 10129 |
ISSN | Citations | PageRank |
0302-9743 | 0 | 0.34 |
References | Authors | |
0 | 5 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Rahil Shahzad | 1 | 0 | 0.68 |
| Shan Gao | 2 | 0 | 0.34 |
| Qian Tao | 3 | 59 | 14.00 |
| Oleh Dzyubachyk | 4 | 0 | 0.34 |
| Rob J. Van Der Geest | 5 | 559 | 56.91 |