SRAD and optical flow based external energy for echocardiograms with primitive shape priors

Authors:
Ali K. Hamou;Mahmoud R. El-Sakka
Affiliations:
Computer Science Department, The University of Western Ontario, London, ON, Canada;Computer Science Department, The University of Western Ontario, London, ON, Canada
Venue:
ICIP'09 Proceedings of the 16th IEEE international conference on Image processing
Year:
2009

Citing 8
Cited 0

Scale-Space and Edge Detection Using Anisotropic Diffusion

IEEE Transactions on Pattern Analysis and Machine Intelligence
Performance of optical flow techniques

International Journal of Computer Vision
3D Cardiac Deformation from Ultrasound Images

MICCAI '99 Proceedings of the Second International Conference on Medical Image Computing and Computer-Assisted Intervention
Space and Time Shape Constrained Deformable Surfaces for 4D Medical Image Segmentation

MICCAI '00 Proceedings of the Third International Conference on Medical Image Computing and Computer-Assisted Intervention
Snakes, shapes, and gradient vector flow

IEEE Transactions on Image Processing
Speckle reducing anisotropic diffusion

IEEE Transactions on Image Processing
Edge detection in ultrasound imagery using the instantaneous coefficient of variation

IEEE Transactions on Image Processing
Carotid ultrasound segmentation using DP active contours

ICIAR'07 Proceedings of the 4th international conference on Image Analysis and Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

Accurate left ventricle border delineation is highly desirable on inherently noisy clinical echocardiograms. Active contour (or snake) is a powerful model-based image segmentation approach. In this work, we propose to use a modified gradient vector flow (GVF) snake to segment noisy echocardiographic image cycles. The first modification is to use a speckle reducing anisotropic diffusion (SRAD) operator to reduce the inherent speckle noise within the image. The second modification is to utilize the movement of the vessels and tissues (identified by means of optical flow analysis) and incorporate it into the external energy of the GVF snake. This will provide the necessary structural information while ignoring static noise prone areas of the image cine. Finally, the incorporation of an iterative priori knowledge process into the proposed solution will retract an expanding curve or correct a caving one when an expected border is occluded by noise. Results are compared with expert-defined segmentations yielding better sensitivity, precision rate and overlap ratio than that of the standard GVF model.