A Fuzzy Approach to Liver Tumor Segmentation with Zernike Moments
In this paper, we present a new segmentation approach
for liver lesions in regions of interest within MRI (Magnetic
Resonance Imaging). This approach, based on a two-cluster Fuzzy CMeans
methodology, considers the parameter variable compactness
to handle uncertainty. Fine boundaries are detected by a local
recursive merging of ambiguous pixels with a sequential forward
floating selection with Zernike moments. The method has been tested
on both synthetic and real images. When applied on synthetic images,
the proposed approach provides good performance, segmentations
obtained are accurate, their shape is consistent with the ground truth,
and the extracted information is reliable. The results obtained on MR
images confirm such observations. Our approach allows, even for
difficult cases of MR images, to extract a segmentation with good
performance in terms of accuracy and shape, which implies that the
geometry of the tumor is preserved for further clinical activities (such
as automatic extraction of pharmaco-kinetics properties, lesion
characterization, etc.).
[1] P. Maji and S. Pal. “Maximum Class Separability for Rough-Fuzzy CMeans
Based Brain MR Image Segmentation,” T. Rough Sets, Vol.9,
pp.114-134, 2008.
[2] “Mathematical Morphology,” (accessed May, 17 2015)
http://www.cs.uu.nl/docs/vakken/ibv/reader/chapter6.pdf
[3] R. Crane, “Simplified Approach to Image Processing,” Prentice Hall,
1997.
[4] J. Russ, “Image Analysis of Food Microstructure,” CRC Press, 2004.
[5] J. Kim, “Image Enhancement for Improving Object Recognition,”
Algorithm and SoC Design for Automotive Vision Systems, Springer,
2014.
[6] S. Roy, H. K. Agarwal, A. Carass, Y. Bai, D. L. Pham, J. L. Prince,
“Fuzzy C-Means with Variable Compactness,” IEEE International
Symposium on Biomedical Imaging, pp. 452-455, 2008.
[7] P. Pudil, J. Novovičová, J. Kittler, “Floating Search Methods in Feature
Selection,” Pattern Recognition Letters, volume 15, pages 1119-1125,
1994.
[8] A. Khotanzan, Y. H. Hong, “Invariant Image Recognition by Zernike
Moments,” IEEE Trans. PAMI, 12:489–497, 1990.
[9] B. Oluleye, A. Leisa, J. Leng, D. Dean, “Zernike Moments and Genetic
Algorithm: Tutorial and Application,” British Journal of Mathematics &
Computer Science 4(15): 2217-2236, 2014.
[10] A. Ali, A. Albouy-Kissi, A. Vacavant, M. Grand-brochier, J. Boire, “A
Novel Fuzzy C-Means Based Defuzzification Approach with an Adapted
Minkowski Distance,” 19th Computer Vision Winter Workshop Zuzana
Kúkelová and Jan Heller (eds.) Křtiny, Czech Republic, 2014.
[11] R. Fang, R. Zabih, A. Raj, T. Chen, “Segmentation of Liver Tumor
Using Efficient Global Optimal Tree Metrics Graph Cuts," In
Abdominal Imaging. Computational and Clinical Applications, 2012.
[12] D. Pham, “Robust Fuzzy Segmentation of Magnetic Resonance Images,
14th IEEE Symposium on Computer-Based Medical Systems, p. 127-
131, 2001. [13] C. Teh and R.T. Chin, “On Image Analysis by the Methods of
Moments,” IEEE Trans. on PAMI, 10 (4). 496-513, 1988.
[14] M. Teague, “Image Analysis via the General Theory of Moments,”
Journal of the Optical Society of America, 70 (8). 920-930, 1980.
[15] J.S. Lipscomb, “A Trainable Gesture Recognizer,” Pattern Recognition,
24 (9). 895-907, 1991.
[16] H. Hse and A. Newton, “Sketched Symbol Recognition Using Zernike
Moments,” In Proceedings of ICPR, 2004.
[17] A. Tahmasbi, F. Saki, and S. B. Shokouhi, "Classification of Benign and
Malignant Masses Based on Zernike Moments," J. Computers in
Biology and Medicine, vol. 41, no. 8, pp. 726-735, 2011.
[18] A. Tahmasbi, F. Saki, H. Aghapanah, and S. B. Shokouhi, "A Novel
Breast Mass Diagnosis System based on Zernike Moments as Shape and
Density Descriptors," In Proc. IEEE, 18th Iranian Conf. on Biomedical
Engineering (ICBME'2011), Tehran, Iran, pp. 100-104, 2011.
[19] M. Grand-Brochier, A. Vacavant, G. Cerutti, K. Bianchi, and L. Tougne,
“Comparative Study of Segmentation Methods for Tree Leaves
Extraction,” In ACM ICVS 2013, Workshop: VIGTA, Saint Petersburg,
Russia, 2013.
[1] P. Maji and S. Pal. “Maximum Class Separability for Rough-Fuzzy CMeans
Based Brain MR Image Segmentation,” T. Rough Sets, Vol.9,
pp.114-134, 2008.
[2] “Mathematical Morphology,” (accessed May, 17 2015)
http://www.cs.uu.nl/docs/vakken/ibv/reader/chapter6.pdf
[3] R. Crane, “Simplified Approach to Image Processing,” Prentice Hall,
1997.
[4] J. Russ, “Image Analysis of Food Microstructure,” CRC Press, 2004.
[5] J. Kim, “Image Enhancement for Improving Object Recognition,”
Algorithm and SoC Design for Automotive Vision Systems, Springer,
2014.
[6] S. Roy, H. K. Agarwal, A. Carass, Y. Bai, D. L. Pham, J. L. Prince,
“Fuzzy C-Means with Variable Compactness,” IEEE International
Symposium on Biomedical Imaging, pp. 452-455, 2008.
[7] P. Pudil, J. Novovičová, J. Kittler, “Floating Search Methods in Feature
Selection,” Pattern Recognition Letters, volume 15, pages 1119-1125,
1994.
[8] A. Khotanzan, Y. H. Hong, “Invariant Image Recognition by Zernike
Moments,” IEEE Trans. PAMI, 12:489–497, 1990.
[9] B. Oluleye, A. Leisa, J. Leng, D. Dean, “Zernike Moments and Genetic
Algorithm: Tutorial and Application,” British Journal of Mathematics &
Computer Science 4(15): 2217-2236, 2014.
[10] A. Ali, A. Albouy-Kissi, A. Vacavant, M. Grand-brochier, J. Boire, “A
Novel Fuzzy C-Means Based Defuzzification Approach with an Adapted
Minkowski Distance,” 19th Computer Vision Winter Workshop Zuzana
Kúkelová and Jan Heller (eds.) Křtiny, Czech Republic, 2014.
[11] R. Fang, R. Zabih, A. Raj, T. Chen, “Segmentation of Liver Tumor
Using Efficient Global Optimal Tree Metrics Graph Cuts," In
Abdominal Imaging. Computational and Clinical Applications, 2012.
[12] D. Pham, “Robust Fuzzy Segmentation of Magnetic Resonance Images,
14th IEEE Symposium on Computer-Based Medical Systems, p. 127-
131, 2001. [13] C. Teh and R.T. Chin, “On Image Analysis by the Methods of
Moments,” IEEE Trans. on PAMI, 10 (4). 496-513, 1988.
[14] M. Teague, “Image Analysis via the General Theory of Moments,”
Journal of the Optical Society of America, 70 (8). 920-930, 1980.
[15] J.S. Lipscomb, “A Trainable Gesture Recognizer,” Pattern Recognition,
24 (9). 895-907, 1991.
[16] H. Hse and A. Newton, “Sketched Symbol Recognition Using Zernike
Moments,” In Proceedings of ICPR, 2004.
[17] A. Tahmasbi, F. Saki, and S. B. Shokouhi, "Classification of Benign and
Malignant Masses Based on Zernike Moments," J. Computers in
Biology and Medicine, vol. 41, no. 8, pp. 726-735, 2011.
[18] A. Tahmasbi, F. Saki, H. Aghapanah, and S. B. Shokouhi, "A Novel
Breast Mass Diagnosis System based on Zernike Moments as Shape and
Density Descriptors," In Proc. IEEE, 18th Iranian Conf. on Biomedical
Engineering (ICBME'2011), Tehran, Iran, pp. 100-104, 2011.
[19] M. Grand-Brochier, A. Vacavant, G. Cerutti, K. Bianchi, and L. Tougne,
“Comparative Study of Segmentation Methods for Tree Leaves
Extraction,” In ACM ICVS 2013, Workshop: VIGTA, Saint Petersburg,
Russia, 2013.
@article{"International Journal of Medical, Medicine and Health Sciences:70564", author = "Abder-Rahman Ali and Antoine Vacavant and Manuel Grand-Brochier and Adélaïde Albouy-Kissi and Jean-Yves Boire", title = "A Fuzzy Approach to Liver Tumor Segmentation with Zernike Moments", abstract = "In this paper, we present a new segmentation approach
for liver lesions in regions of interest within MRI (Magnetic
Resonance Imaging). This approach, based on a two-cluster Fuzzy CMeans
methodology, considers the parameter variable compactness
to handle uncertainty. Fine boundaries are detected by a local
recursive merging of ambiguous pixels with a sequential forward
floating selection with Zernike moments. The method has been tested
on both synthetic and real images. When applied on synthetic images,
the proposed approach provides good performance, segmentations
obtained are accurate, their shape is consistent with the ground truth,
and the extracted information is reliable. The results obtained on MR
images confirm such observations. Our approach allows, even for
difficult cases of MR images, to extract a segmentation with good
performance in terms of accuracy and shape, which implies that the
geometry of the tumor is preserved for further clinical activities (such
as automatic extraction of pharmaco-kinetics properties, lesion
characterization, etc.).", keywords = "Defuzzification, floating search, fuzzy clustering,
Zernike moments.", volume = "9", number = "7", pages = "559-6", }
