Projects:SegmentationEpicardialWall
Back to Georgia Tech Algorithms
3D View of the Segmentation of Endocardial Wall
2D View
Key Investigators
- Behnood Gholami, Yi Gao, Wassim M. Haddad, and Allen Tannenbaum, Georgia Tech
- Rob MacLeod, and Josh Blauer, University of Utah
Description
Atrial fibrillation, a cardiac arrhythmia characterized by unsynchronized electrical activity in the atrial chambers of the heart, is a rapidly growing problem in modern societies. Electrical cardioversion and antiarrhythmic drugs are used to manage this condition, but suffer from low success rates and involve major side effects. In an alternative treatment, known as catheter ablation, specific parts of the left atrium are targeted for radio frequency ablation using an intracardiac catheter. Application of radio frequency energy to the cardiac tissue causes thermal injury (lesions), which in turn results into scar tissue. Successful ablation can eliminate, or isolate, the problematic sources of electrical activity and effectively cure atrial fibrillation.
Magnetic resonance imaging (MRI) has been used for both pre- and and post-ablation assessment of the atrial wall. MRI can aid in selecting the right candidate for the ablation procedure and assessing post-ablation scar formations. Image processing techniques can be used for automatic segmentation of the atrial wall, which facilitates an accurate statistical assessment of the region. As a first step towards the general solution to the computer-assisted segmentation of the left atrial wall, in this paper we propose a shape-based image segmentation framework to segment the endocardial wall of the left atrium.
Our Approach
Our proposed method is composed of shape registration, shape learning, and image segmentation. Given a training set of binary images corresponding to the segmentations of the epicardial wall of the left atrium, each shape is registered to a fixed arbitrary shape in the training set using a mean-square error registration scheme. Next, we use principal component analysis to learn the shapes. Finally, the shape information available from the learning stage is used to segment the epicardial wall, where the segmentation is driven by local regional statistics. We trained the algorithm on 15 segmentations of the epicardial wall of the left atrium performed by a human expert. The first 6 dominant principal components were used in the statistical shape learning. The algorithm was then used to segment the epicardial wall, as shown in the figure.
Publications
In Press
Y. Gao, B. Gholami, R. S. MacLeod, J, Blauer, W. M. Haddad, and A. R. Tannenbaum, Segmentation of the Epicardial Wall of the Left Atrium Using Statistical Shape Learning and Local Curve Statistics, SPIE Medical Imaging 2010
Key Investigators
Georgia Tech: Behnood Gholami, Yi Gao, Wassim Haddad, and Allen Tannenbaum University of Utah: Rob MacLeod, and Josh Blauer
