A Novel Approach For 3D Surface Reconstruction Of LV Endocardium Based On SPCNN And MC

DOI

10.2991/iceeecs-16.2016.12How to use a DOI?

Keywords

3D surface reconstruction, Left ventricle, Cubic Spline Interpolation, SPCNN segmentation, MC

Abstract

The auxiliary diagnosis based on three-dimensional (3D) modeling techniques can improve the diagnosis efficiency of heart disease. In this paper, we propose a new method for the 3D modeling of left ventricle. The novel approach incorporatestwo key points: one is LV endocardium segmentation via the simplified pulse-coupled neural network (SPCNN), and the other is interpolation between slices. To ensure the accuracy of modeling, we firstly use the cubic spline interpolation to increase the resolution between slices. Then, the endocardial contours of all slices are automatically obtained based on SPCNN. Finally, Marching Cubes (MC) algorithm is implemented to reconstruct the LV endocardial surface. We test the new method on four setsof 3D data and obtain effective results. The APD and ODM are respectively 1.7950mm and 91.34% for the endocardium segmentation, which makes the surface reconstruction of LV endocardium accurate andsmooth.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Ruoming Lei
AU  - Yurun Ma
AU  - Kemin Wang
AU  - Yide Ma
PY  - 2016/12
DA  - 2016/12
TI  - A Novel Approach For 3D Surface Reconstruction Of LV Endocardium Based On SPCNN And MC
BT  - Proceedings of the 2016 4th International Conference on Electrical & Electronics Engineering and Computer Science (ICEEECS 2016)
PB  - Atlantis Press
SP  - 53
EP  - 60
SN  - 2352-538X
UR  - https://doi.org/10.2991/iceeecs-16.2016.12
DO  - 10.2991/iceeecs-16.2016.12
ID  - Lei2016/12
ER  -