Segmentation of the Left Ventricle in Myocardial Perfusion SPECT Using Active Shape Model

Authors:
Wooi-Haw Tan;Rosli Besar
Affiliations:
Center for Multimedia Security and Signal Processing, Multimedia University, Persiaran Multimedia, Cyberjaya, Malaysia 63100;Center for Multimedia Security and Signal Processing, Multimedia University, Persiaran Multimedia, Cyberjaya, Malaysia 63100
Venue:
IVIC '09 Proceedings of the 1st International Visual Informatics Conference on Visual Informatics: Bridging Research and Practice
Year:
2009

Citing 2
Cited 0

Active shape models—their training and application

Computer Vision and Image Understanding
Snakes, shapes, and gradient vector flow

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In the quantification of myocardial perfusion SPECT (MPS), numerous processes are involved. Automation is desired as it will considerably reduce the laboriousness of the underlying tasks. In this paper, we propose a segmentation scheme for the delineation of left ventricle (LV) using the Active Shape Models. Our scheme will reduce the labour-intensiveness in MPS quantification, while still allowing interactive guidance from the medical experts. The proposed scheme has been applied on clinical MPS tomograms in which it has successfully delineated the LV in 94% of the test data. In addition, it has also shown to be more suitable for LV segmentation than the rivaling Active Contour Model.