Magnetic-resonance image segmentation based on improved variable weight multi-resolution Markov random field in undecimated complex wavelet domain

doi:10.1088/1674-1056/abe1a2

中国物理B ›› 2021, Vol. 30 ›› Issue (7): 78703-078703.doi: 10.1088/1674-1056/abe1a2

Magnetic-resonance image segmentation based on improved variable weight multi-resolution Markov random field in undecimated complex wavelet domain

Hong Fan(范虹)^1,†, Yiman Sun(孙一曼)¹, Xiaojuan Zhang(张效娟)², Chengcheng Zhang(张程程)¹, Xiangjun Li(李向军)³, and Yi Wang(王乙)⁴

1 School of Computer Science, Shaanxi Normal University, Xi'an 710062, China;
2 School of Computer Science, Qinghai Normal University, Xining 810003, China;
3 School of Information Engineering, Xi'an University, Xi'an 710065, China;
4 Department of Biomedical Engineering, Cornell University, Ithaca, NY 14853, USA

收稿日期:2020-10-21 修回日期:2021-01-14 接受日期:2021-02-01 出版日期:2021-06-22 发布日期:2021-06-22
通讯作者: Hong Fan E-mail:fanhong@snnu.edu.cn
基金资助:
Project supported by the National Natural Science Foundation of China (Grant No. 11471004), and the Key Research and Development Program of Shaanxi Province, China (Grant No. 2018SF-251).

Magnetic-resonance image segmentation based on improved variable weight multi-resolution Markov random field in undecimated complex wavelet domain

Hong Fan(范虹)^1,†, Yiman Sun(孙一曼)¹, Xiaojuan Zhang(张效娟)², Chengcheng Zhang(张程程)¹, Xiangjun Li(李向军)³, and Yi Wang(王乙)⁴

1 School of Computer Science, Shaanxi Normal University, Xi'an 710062, China;
2 School of Computer Science, Qinghai Normal University, Xining 810003, China;
3 School of Information Engineering, Xi'an University, Xi'an 710065, China;
4 Department of Biomedical Engineering, Cornell University, Ithaca, NY 14853, USA

Received:2020-10-21 Revised:2021-01-14 Accepted:2021-02-01 Online:2021-06-22 Published:2021-06-22
Contact: Hong Fan E-mail:fanhong@snnu.edu.cn
Supported by:
Project supported by the National Natural Science Foundation of China (Grant No. 11471004), and the Key Research and Development Program of Shaanxi Province, China (Grant No. 2018SF-251).

摘要/Abstract

摘要： To solve the problem that the magnetic resonance (MR) image has weak boundaries, large amount of information, and low signal-to-noise ratio, we propose an image segmentation method based on the multi-resolution Markov random field (MRMRF) model. The algorithm uses undecimated dual-tree complex wavelet transformation to transform the image into multiple scales. The transformed low-frequency scale histogram is used to improve the initial clustering center of the K-means algorithm, and then other cluster centers are selected according to the maximum distance rule to obtain the coarse-scale segmentation. The results are then segmented by the improved MRMRF model. In order to solve the problem of fuzzy edge segmentation caused by the gray level inhomogeneity of MR image segmentation under the MRMRF model, it is proposed to introduce variable weight parameters in the segmentation process of each scale. Furthermore, the final segmentation results are optimized. We name this algorithm the variable-weight multi-resolution Markov random field (VWMRMRF). The simulation and clinical MR image segmentation verification show that the VWMRMRF algorithm has high segmentation accuracy and robustness, and can accurately and stably achieve low signal-to-noise ratio, weak boundary MR image segmentation.

关键词: undecimated dual-tree complex wavelet, MR image segmentation, multi-resolution Markov random field model

Abstract: To solve the problem that the magnetic resonance (MR) image has weak boundaries, large amount of information, and low signal-to-noise ratio, we propose an image segmentation method based on the multi-resolution Markov random field (MRMRF) model. The algorithm uses undecimated dual-tree complex wavelet transformation to transform the image into multiple scales. The transformed low-frequency scale histogram is used to improve the initial clustering center of the K-means algorithm, and then other cluster centers are selected according to the maximum distance rule to obtain the coarse-scale segmentation. The results are then segmented by the improved MRMRF model. In order to solve the problem of fuzzy edge segmentation caused by the gray level inhomogeneity of MR image segmentation under the MRMRF model, it is proposed to introduce variable weight parameters in the segmentation process of each scale. Furthermore, the final segmentation results are optimized. We name this algorithm the variable-weight multi-resolution Markov random field (VWMRMRF). The simulation and clinical MR image segmentation verification show that the VWMRMRF algorithm has high segmentation accuracy and robustness, and can accurately and stably achieve low signal-to-noise ratio, weak boundary MR image segmentation.

Key words: undecimated dual-tree complex wavelet, MR image segmentation, multi-resolution Markov random field model

中图分类号: (Image analysis)

87.57.N-

87.61.-c (Magnetic resonance imaging) 87.63.lm (Image enhancement)

Hong Fan(范虹), Yiman Sun(孙一曼), Xiaojuan Zhang(张效娟), Chengcheng Zhang(张程程), Xiangjun Li(李向军), and Yi Wang(王乙). Magnetic-resonance image segmentation based on improved variable weight multi-resolution Markov random field in undecimated complex wavelet domain[J]. 中国物理B, 2021, 30(7): 78703-078703.

参考文献

[1] Bao S L, Du J and Gao S 2013 Acta Phys. Sin. 62 088701 (in Chinese)
[2] Mann R M, Kuhl C K, Kinkel K and Boetes C 2008 Eur. Radiol. 18 1307
[3] Jeevakala S, Brintha T A and Rangasami R 2018 Biomed. Signal Process. Control 39 117
[4] Panigrahi L, Verma K and Singh B K 2019 Expert Systems with Applications 115 486
[5] Zhang J, Saha A, Zhu Z and Mazurowski M A 2018 IEEE Trans. Med. Imaging 38 435
[6] Ahmadvand A and Daliri M R 2015 Appl. Math. Comput. 256 808
[7] Geman S and Geman D 1984 IEEE Trans. Pattern Analysis and Machine Intelligence 6 721
[8] Azmi R and Norozi N 2011 J. Med. Signals and Sensors 1 156
[9] Ribes S, Didierlaurent D, Decoster N, Gonneau E, Risser L, Feillel V and Caselles O 2014 IEEE Trans. Med. Imaging 33 1986
[10] Goubalan S R, Goussard Y and Maaref H 2016 2016 IEEE International Conference on Image Processing (ICIP) September 25-28, 2016, Phoenix, USA, p. 2653
[11] Lee G G and Chen C H 1996 1996 IEEE Nuclear Science Symposium. Conference Record November 2-9, 1996, Anaheim, USA, p. 1937
[12] Wang L and Liu J 1999 Pattern Recognition Lett. 20 171
[13] Liu G and Guo T 2010 2010 6th International Conference on Wireless Communications Networking and Mobile Computing (WiCOM), September 23-25, 2010, Chengdu, China, p. 1
[14] Tang W, Zhang C and Zou H 2013 Intelligent Computing Theories and Technology July, 2013, Berlin, Heidelberg, p. 56
[15] Hill P, Achim A and Bull D 2012 2012 19th IEEE International Conference on Image Processing September 30-October 3, 2012, Orlando, USA, p. 1205
[16] Selesnick I W, Baraniuk R G and Kingsbury N C 2005 IEEE Signal Process. Mag. 22 123
[17] Kaur G and Vashist S 2017 International Journal of Computer Applications 167 26
[18] Kingsbury N 1999 Philosophical Transactions Mathematical Physical and Engineering Sciences 357 2543
[19] Jain A K 2010 Pattern Recognition Lett. 31 651
[20] Celeux G, Forbes F and Peyrard N 2003 Pattern Recognition 36 131
[21] Fan H, Zhu Y C, Wang F M and Zhang X M 2014 Acta Phys. Sin 63 392 (in Chinese)
[22] Web B 2004 Simulated Brain Database McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill
[23] Turaga D S, Chen Y and Caviedes J 2004 Signal Processing: Image Commun. 19 173
[24] Dice L R 1945 Ecology 26 297
[25] Menze B H, Jakab A, Bauer S et al. 2015 IEEE Trans. Med. Imaging 34 1993
[26] Bakas S, Akbari H, Sotiras A, Bilello M, Rozycki M, Kirby J S, Freymann J B, Farahani K and Davatzikos C 2017 Sci. Data 4 170117
[27] Bakas S, Reyes M, Jakab A et al. 2018 arXiv:1811.02629v3 [cs.CV]

Magnetic-resonance image segmentation based on improved variable weight multi-resolution Markov random field in undecimated complex wavelet domain

Magnetic-resonance image segmentation based on improved variable weight multi-resolution Markov random field in undecimated complex wavelet domain

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 9

Metrics

本文评价

推荐阅读 0

[1]	Zhi-Li Wang(王志立), Zi-Han Chen(陈子涵), Yao Gu(顾瑶), Heng Chen(陈恒), and Xin Ge(葛昕). Investigations of moiré artifacts induced by flux fluctuations in x-ray dark-field imaging[J]. 中国物理B, 2023, 32(3): 38704-038704.
[2]	Li-Ming Zhao(赵立明), Tian-Xiang Wang(王天祥), Run-Kang Ma(马润康), Yao Gu(顾瑶), Meng-Si Luo(罗梦丝), Heng Chen(陈恒), Zhi-Li Wang(王志立), and Xin Ge(葛昕). Analysis of refraction and scattering image artefacts in x-ray analyzer-based imaging[J]. 中国物理B, 2023, 32(2): 28701-028701.
[3]	. [J]. 中国物理B, 2021, 30(1): 18701-.
[4]	张选平, 郭瑞, 陈恒伟, 赵仲孟, 王嘉寅. Efficient image encryption scheme with synchronous substitution and diffusion based on double S-boxes[J]. 中国物理B, 2018, 27(8): 80701-080701.
[5]	张洁敏, 黄晓林, 管群, 刘铁兵, 李萍, 肇莹, 刘红星. Some regularity on how to locate electrodes for higher fECG SNRs[J]. , 2015, 24(3): 38702-038702.
[6]	唐发宽, 王倩, 华宁, 唐雪正, 陆宏, 马平. Forward and inverse problem for cardiac magnetic field and electric potential using two boundary element methods[J]. 中国物理B, 2010, 19(12): 120601-120601.
[7]	贾鹏翔, 张峰, 闫镔, 包尚联. Scattering correction method for panel detector based cone beam computed tomography system[J]. 中国物理B, 2010, 19(8): 87802-087802.
[8]	王林元, 李磊, 闫镔, 江成顺, 王浩宇, 包尚联. An algorithm for computed tomography image reconstruction from limited-view projections[J]. 中国物理B, 2010, 19(8): 88106-088106.
[9]	董光炯, 韩汝珊, 黄畇, KAORU SAKATANI, 庄逢源. TOMOGRAPHY FORMULA FOR BIOCHEMICAL IMAGING OF THIN TISSUE WITH DIFFUSE-PHOTON DENSITY WAVES[J]. 中国物理B, 1999, 8(12): 938-943.