DOI: 10.3724/SP.J.1001.2009.00030

Journal of Software (软件学报) 2009/20:1 PP.30-40

Texture Classification and Shape Statistics Variational Approach for Segmentation of Left Ventricle Tagged MR Images

Segmentation of left ventricle tagged MR images is the basis of ventricular motion reconstruction. In left ventricle tagged MR images, the boundaries are often obscured or corrupted by the tag lines and region inhomogeneity as well as the existence of papillary muscles. These factors increase the difficulty of segmenting the inner and outer contour of left ventricle precisely. This paper introduces texture classification information and shape statistical knowledge into the Mumford-Shah model and presents an improved texture classification and shape statistics variational approach for the segmentation of inner and outer contour of left ventricle. It uses the output of support vector machine (SVM) classifier relying on S filter banks to construct a new region-based image energy term. This approach can overcome the influence of tag lines because it makes use of the supervised classification strategy. The introduction of shape statistics can improve the segmentation with broken boundaries. Segmentation results of an entire cardiac period on an identical image layer and a comparison of mean absolute distance analysis between contours generated by this approach and that generated by hand demonstrate that this method can achieve a higher segmentation precision and a better stability than other various approaches.

Key words:Mumford-Shah model,S filter bank,SVM (support vector machine),shape statistics,left ventricle segmentation

ReleaseDate:2014-07-21 14:29:10

Funds:Supported by the National Natural Science Foundation of China under Grant No.60773172 (国家自然科学基金); the Natural Science Foundation of Jiangsu Province of China under Grant No.BK2006704-2 (江苏省自然科学基金)

[1] Metaxas D, Chen T, Huang X, Axel L. Cardiac segmentation from MRI-tagged and CT images. In: Bojkovic Z, ed. Proc. of the 8th WSEAS Int'l Conf. on Computers, Special Session on Imaging and Image Processing of Dynamic Processes in Biology and Medicine WSEAS Trans. Athens: WSEAS Press, 2004. 587-592.

[2] Guttman M, Prince J, McVeigh E. Tag and contour detection in tagged MR images of the left ventricle. IEEE Trans. on Medial Imaging, 1994,13(1):74-88.

[3] Chen Q, Zhou ZM, Qu YG, Heng PA, Xia DS. Auto segmentation of the left ventricle MR images. Chinese Journal of Computers, 2005,28(6):991-999 (in Chinese with English abstract).陈强,周则明,屈颖歌,王平安,夏德深.左心室核磁共振图像的自动分割.计算机学报,2005,28(6):991-999.

[4] Qian Z, Metaxas D, Axel L. A learning framework for the automatic and accurate segmentation of cardiac tagged MRI images. In: Liu YX, Jiang TZ, Zhang CS, eds. Computer Vision for Biomedical Image Applications Workshop. Beijing: Springer-Verlag, 2005, 93-102.

[5] Histace A, Cavaro-Menard C, Vigouroux B. Tagged cardiac MRI: Detection of myocardial boundaries by texture analysis. In: SuviSoft Oy Ltd., ed. Int'l Conf. on Image Processing. Barcelona: IEEE Computer Society Press, 2003,2:1061-1064.

[6] You JJ, Wang PA, Xia DS. Tagged cardiac magnetic resonance images segmentation using texture information. Chinese Journal of Image and Graphics, 2007,12(9):1621-1626 (in Chinese with English abstract).尤建洁,王平安,夏德深.利用纹理信息的带标记线心脏核磁共振图像分割.中国图象图形学报,2007,12(9):1621-1626.

[7] Cremers D, Tischhäuser F, Weickert J, Schnörr C. Diffusion snakes: Introducing statistical shape knowledge into the Mumford-Shah functional. Int'l Journal of Computer Vision, 2002,50(3):295-313.

[8] Chen Q, Zhou ZM, Tang M, Heng PA, Shen XD. Shape statistics variational approach for the outer contour segmentation of left ventricle MR images. IEEE Trans. on Information Technology in BioMedicine, 2006,10(3):588-597.

[9] Varma M, Zisserman A. A statistical approach to texture classification from single images. Int'l Journal of Computer Vision, 2005, 62(1-2):61-81.

[10] Boser BE, Guyon IM, Vapnik VN. A training algorithm for optimal margin classifiers. In: Haussler D, ed. Proc. of the 5th Annual ACM Workshop on COLT. Pittsburgh: ACM Press, 1992. 144-152.

[11] Kim KI, Jung K, Park SH, Kim HJ. Support vector machines for texture classification. IEEE Trans. on Pattern Analysis and Machine Intelligence, 2002,24(11):1542-1550.

[12] Hu ZP, Zhang Y. Improved Mumford-Shah active contour image segmentation based on support vector machine energy representation. Acta Electronica Sinica, 2006,34(5):930-933 (in Chinese with English abstract).胡正平,张晔.基于SVM能量模型的改进主动轮廓图像分割算法研究.电子学报,2006,34(5):930-933.

[13] Cootes TF, Taylor CJ, Cooper DH, Graham J. Active shape models-Their training and application. Computer Vision and Image Understanding, 1995,61(1):38-59.

[14] Leventon ME, Grimson WEL, Faugeras O. Statistical shape influence in geodesic active contours. In: Kriegman DJ, ed. IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, Vol.1. Hilton Head Island: IEEE Computer Society Press, 2000. 1316-1323.

[15] Cremers D, Kohlberger T, Schnörr C. Shape statistics in kernel space for variational image segmentation. Pattern Recognition, 2003,36(9):1929-1943.

[16] Sonka M, Hlavac V, Boyle R. Image Process, Analysis, and Machine Vision. 2th ed., Beijing: Posts & Telecom Press, 2002. 380-389.

[17] Werman M, Weinshall D. Similarity and affine invariant distances between 2D point sets. IEEE Trans. Pattern Analysis Machine Intelligence, 1995,17(8):810-814.

[18] Mikić I, Krucinski S, Thomas JD. Segmentation and tracking in echocardiographic sequences: Active contours guided by optical flow estimates. IEEE Trans. on Medical Imaging, 1998,17(2):274-283.