Introducing Shape Constraint Via Legendre Moments in a Variational Framework for Cardiac Segmentation on Non-contrast CT Images

In thoracic radiotherapy, some organs should be considered with care and protected from undesirable radiation. Among these organs, the heart is one of the most critical to protect. Its segmentation from routine CT scans provides valuable information to assess its position and shape. In this paper, we present a novel variational segmentation method for extracting the heart on non-contrast CT images. To handle the low image contrast around the cardiac borders, we propose to integrate shape constraints using Legendre moments and adding an energy term in the functional to be optimized. Results for whole heart segmentation in non-contrast CT images are presented and comparisons are performed with manual segmentations.

[1]  A. Morenoa,et al.  Using anatomical knowledge expressed as fuzzy constraints to segment the heart in CT images , 2008 .

[2]  Benoit Mory,et al.  Fuzzy Region Competition: A Convex Two-Phase Segmentation Framework , 2007, SSVM.

[3]  Rachid Deriche,et al.  A Review of Statistical Approaches to Level Set Segmentation: Integrating Color, Texture, Motion and Shape , 2007, International Journal of Computer Vision.

[4]  Michel Barlaud,et al.  Combining shape prior and statistical features for active contour segmentation , 2004, IEEE Transactions on Circuits and Systems for Video Technology.

[5]  Isabelle Bloch,et al.  Using anatomical knowledge expressed as fuzzy constraints to segment the heart in CT images , 2008, Pattern Recognit..

[6]  Dominique Hasboun,et al.  Multi-object Deformable Templates Dedicated to the Segmentation of Brain Deep Structures , 1998, MICCAI.

[7]  Mila Nikolova,et al.  Finding the global minimum for binary image restoration , 2005, IEEE International Conference on Image Processing 2005.

[8]  Olivier D. Faugeras,et al.  Statistical shape influence in geodesic active contours , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[9]  Christophe Odet,et al.  Shape prior criterion based on Tchebichef moments in variational region growing , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).

[10]  Khalid M. Hosny,et al.  Exact Legendre moment computation for gray level images , 2007, Pattern Recognit..

[11]  Tony F. Chan,et al.  Active contours without edges , 2001, IEEE Trans. Image Process..

[12]  Fabrice Heitz,et al.  Affine-invariant geometric shape priors for region-based active contours , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Rachid Deriche,et al.  Geodesic active contours for supervised texture segmentation , 1999, Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149).

[14]  Olivier Ecabert,et al.  Automatic Model-Based Segmentation of the Heart in CT Images , 2008, IEEE Transactions on Medical Imaging.

[15]  M. Teague.  Image analysis via the general theory of moments , 1980 .