Segmentation of Breast Lesions in Ultrasound Images Using Spatial Fuzzy Clustering and Structure Tensors

Segmentation in ultrasound images is challenging due to the interference from speckle noise and fuzziness of boundaries. In this paper, a segmentation scheme using fuzzy c-means (FCM) clustering incorporating both intensity and texture information of images is proposed to extract breast lesions in ultrasound images. Firstly, the nonlinear structure tensor, which can facilitate to refine the edges detected by intensity, is used to extract speckle texture. And then, a spatial FCM clustering is applied on the image feature space for segmentation. In the experiments with simulated and clinical ultrasound images, the spatial FCM clustering with both intensity and texture information gets more accurate results than the conventional FCM or spatial FCM without texture information.

• Yan Xu
• Toshihiro Nishimura

• fuzzy c-means
• spatial information
• structure tensor
• ultrasound image segmentation

[1] Y. L. Huang, D. R. Chen, Y. R. Jiang, S. J. Kuo, H. K. Wu, and W.
K. Moon, "Computer-aided diagnosis using morphological features for
classifying breast lesions on ultrasound," Ultrasound in Obstetrics and
Gynecology, vol. 32, pp. 565-572, Sep 2008.
[2] K. Drukker, C. A. Sennett, and M. L. Giger, "Automated Method for
Improving System Performance of Computer-Aided Diagnosis in Breast
Ultrasound," Medical Imaging, IEEE Transactions on, vol. 28, pp. 122-
128, 2009.
[3] J. Weickert. Anisotropic Diffusion in Image Processing. Teubner,
Stuttgart, 1998.
[4] J. C. Bezdek, Pattern Recognition with Fuzzy Objective Function Algorithms.
New York: Plenum Press, 1981.
[5] D. L. Pham, "Spatial Models for Fuzzy Clustering." Computer Vision and
Image Understanding, vol. 84, pp. 285-297, 2001.
[6] A. W. C. Liew, S. H. Leung, and W. H. Lau, "Fuzzy image clustering incorporating
spatial continuity," Iee Proceedings-Vision Image and Signal
Processing, vol. 147, pp. 185-192, Apr 2000.
[7] P. Perona and J. Malik, "Scale-space and edge detection using anisotropic
diffusion," Pattern Analysis and Machine Intelligence, IEEE Transactions
on, vol. 12, pp. 629-639, 1990.
[8] T. Brox, J. Weickert, B. Burgeth, and P. Mrazek, "Nonlinear structure
tensors," Image and Vision Computing, vol. 24, pp. 41-55, Jan 2006.
[9] K. S. Chuang, H. L. Tzeng, S. Chen, J. Wu, and T. J. Chen,
"Fuzzy c-means clustering with spatial information for image segmentation,"
Computerized Medical Imaging and Graphics, vol. 30, pp. 9-15,
2006.
[10] M. Kakar and D. R. Olsen, "Automatic segmentation and recognition
of lungs and lesion from CT scans of thorax," Computerized Medical
Imaging and Graphics, vol. 33, pp. 72-82, 2009.
[11] D. Alexander, J. C. Gee, and R. Bajcsy, "Similarity measure for
matching diffusion tensor images," in British Machine Vision Conference
University of Bottingham, 1999.
[12] J. A. Jensen and S. Nikolov,"Fast simulation of ultrsound images," in
IEEE Ultrason. Symp., 2000, pp. 1721-1724.