Esophagus Segmentation from 3D CT Data Using Skeleton Prior-Based Graph Cut

The segmentation of organs at risk in CT volumes is a prerequisite for radiotherapy treatment planning. In this paper, we focus on esophagus segmentation, a challenging application since the wall of the esophagus, made of muscle tissue, has very low contrast in CT images. We propose in this paper an original method to segment in thoracic CT scans the 3D esophagus using a skeleton-shape model to guide the segmentation. Our method is composed of two steps: a 3D segmentation by graph cut with skeleton prior, followed by a 2D propagation. Our method yields encouraging results over 6 patients.

[1]  Geoffrey G. Zhang,et al.  Semi-automated CT segmentation using optic flow and Fourier interpolation techniques , 2006, Comput. Methods Programs Biomed..

[2]  Tao Zhang,et al.  Interactive graph cut based segmentation with shape priors , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[3]  Timothy F. Cootes,et al.  Active Shape Models-Their Training and Application , 1995, Comput. Vis. Image Underst..

[4]  Brian B. Avants,et al.  Adaptive graph cuts with tissue priors for brain MRI segmentation , 2006, 3rd IEEE International Symposium on Biomedical Imaging: Nano to Macro, 2006..

[5]  Ying Bai,et al.  Probabilistic minimal path for automated esophagus segmentation , 2006, SPIE Medical Imaging.

[6]  W. Eric L. Grimson,et al.  A shape-based approach to the segmentation of medical imagery using level sets , 2003, IEEE Transactions on Medical Imaging.

[7]  Marie-Pierre Jolly,et al.  Interactive Graph Cuts for Optimal Boundary and Region Segmentation of Objects in N-D Images , 2001, ICCV.

[8]  Vladimir Kolmogorov,et al.  An experimental comparison of min-cut/max- flow algorithms for energy minimization in vision , 2001, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Frank Deinzer,et al.  Esophagus Segmentation by Spatially-Constrained Shape Interpolation , 2008, Bildverarbeitung für die Medizin.

[10]  Stephen M. Pizer,et al.  Medial Models of Populations of Nearly Tubular Objects , 2009 .

[11]  Dorin Comaniciu,et al.  Fast Automatic Segmentation of the Esophagus from 3D CT Data Using a Probabilistic Model , 2009, MICCAI.

[12]  Marie-Pierre Jolly,et al.  Interactive graph cuts for optimal boundary & region segmentation of objects in N-D images , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[13]  Guido Gerig,et al.  Towards a shape model of white matter fiber bundles using diffusion tensor MRI , 2004, 2004 2nd IEEE International Symposium on Biomedical Imaging: Nano to Macro (IEEE Cat No. 04EX821).

[14]  George Starkschall,et al.  Semiautomated four-dimensional computed tomography segmentation using deformable models. , 2005, Medical physics.