A modified intuitionistic fuzzy c-means clustering approach to segment human brain MRI image

Abstract

Fuzzy c-means (FCM) is one of the prominent method utilized for medical image segmentation. In literature intuitionistic fuzzy c-means (IFCM) is suggested which is based on intuitionistic fuzzy sets (IFSs) theory to handle uncertainty and vagueness associated with real data. The objective function of which is defined using the hesitation degree along with membership degree. However, instead of solving the objective function analytically, the approximate solution is obtained using FCM. In this paper, we have proposed a modified intuitionistic fuzzy c-means algorithm (MIFCM) and solved analytically the objective function of the MIFCM method using Lagrange method of undetermined multiplier. To incorporate hesitation degree, two parametric intuitionistic fuzzy complements namely Sugeno’s negation function and Yager’s negation function are investigated. The performance of the MIFCM method is compared with three intuitionistic fuzzy clustering methods and the FCM on two publicly available MRI dataset and a synthetic dataset. The performance measures (average segmentation accuracy, dice score, jaccard score, false negative ratio and false positive ratio) are used to compare the performance of the MIFCM method with three variants of intuitionistic fuzzy clustering methods and the FCM. Experimental results demonstrate the superior performance of the MIFCM method over others.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3

Notes

  1. 1.

    BrainWeb [online], available: http://www.brainweb.bic.mni.mcgill.ca/brainweb.

  2. 2.

    IBSR [online], available: http://www.cma.mgh.harvard.edu/ibsr/

  3. 3.

    Brain Extraction Tool (BET) [online], available: http://www.fmrib.ox.ac.uk/fsl/.

References

  1. 1.

    Alipour S, Shanbehzadeh J (2014) Fast automatic medical image segmentation based on spatial kernel fuzzy c-means on level set method. Mach Vis Appl 25 (6):1469–1488

    Article  Google Scholar 

  2. 2.

    Atanassov KT (1986) Intuitionistic fuzzy sets. Fuzzy Sets Syst 20(1):87–96

    MathSciNet  MATH  Article  Google Scholar 

  3. 3.

    Atanassov KT (2003) Intuitionistic fuzzy sets: past, present and future. In: EUSFLAT conference, pp 12–19

  4. 4.

    Balafar MA, Ramli AR, Saripan MI, Mashohor S (2010) Review of brain mri image segmentation methods. Artif Intell Rev 33(3):261–274

    Article  Google Scholar 

  5. 5.

    Benaichouche A, Oulhadj H, Siarry P (2013) Improved spatial fuzzy c-means clustering for image segmentation using pso initialization, mahalanobis distance and post-segmentation correction. Digital Signal Process 23(5):1390–1400

    MathSciNet  Article  Google Scholar 

  6. 6.

    Bezdek JC (1981) Objective Function Clustering. In: Pattern recognition with fuzzy objective function algorithms. Springer, pp 43–93

  7. 7.

    Bustince H, Kacprzyk J, Mohedano V (2000) Intuitionistic fuzzy generators application to intuitionistic fuzzy complementation. Fuzzy Sets Syst 114(3):485–504

    MathSciNet  MATH  Article  Google Scholar 

  8. 8.

    Chaira T (2011) A novel intuitionistic fuzzy c means clustering algorithm and its application to medical images. Appl Soft Comput 11(2):1711–1717

    Article  Google Scholar 

  9. 9.

    Chen X, Nguyen BP, Chui CK, Ong SH (2016) Automated brain tumor segmentation using kernel dictionary learning and superpixel-level features. In: 2016 IEEE international conference on systems, man, and cybernetics (SMC). IEEE, pp 002,547–002,552

  10. 10.

    Chuang KS, Tzeng HL, Chen S, Wu J, Chen TJ (2006) Fuzzy c-means clustering with spatial information for image segmentation. Comput Med Imaging Graph 30(1):9–15

    Article  Google Scholar 

  11. 11.

    Cocosco CA, Kollokian V, Kwan RKS, Pike GB, Evans AC (1997) Brainweb: online interface to a 3d mri simulated brain database. In: NeuroImage. Citeseer

  12. 12.

    Derrac J, García S, Molina D, Herrera F (2011) A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms. Swarm Evol Comput 1(1):3–18

    Article  Google Scholar 

  13. 13.

    Friedman M (1937) The use of ranks to avoid the assumption of normality implicit in the analysis of variance. J Am Stat Assoc 32(200):675–701

    MATH  Article  Google Scholar 

  14. 14.

    Huang CW, Lin KP, Wu MC, Hung KC, Liu GS, Jen CH (2015) Intuitionistic fuzzy c-means clustering algorithm with neighborhood attraction in segmenting medical image. Soft Comput 19(2):459–470

    Article  Google Scholar 

  15. 15.

    Iakovidis D, Pelekis N, Kotsifakos E, Kopanakis I (2008) Intuitionistic fuzzy clustering with applications in computer vision. In: Advanced concepts for intelligent vision systems. Springer, pp 764–774

  16. 16.

    Iman RL, Davenport JM (1980) Approximations of the critical region of the fbietkan statistic. Communications in Statistics-Theory and Methods 9(6):571–595

    MATH  Article  Google Scholar 

  17. 17.

    Ji ZX, Sun QS, Xia DS (2014) A framework with modified fast fcm for brain mr images segmentation (retraction of vol 44, pg 999, 2011). Pattern Recogn 47 (12):3979–3979

    Article  Google Scholar 

  18. 18.

    Kannan S, Devi R, Ramathilagam S, Takezawa K (2013) Effective fcm noise clustering algorithms in medical images. Comput Biol Med 43(2):73–83

    Article  Google Scholar 

  19. 19.

    Krinidis S, Chatzis V (2010) A robust fuzzy local information c-means clustering algorithm. IEEE Trans Image Process 19(5):1328–1337

    MathSciNet  MATH  Article  Google Scholar 

  20. 20.

    Krishnapuram R, Keller JM (1993) A possibilistic approach to clustering. IEEE Trans Fuzzy Syst 1(2):98–110

    Article  Google Scholar 

  21. 21.

    Li C, Huang R, Ding Z, Gatenby JC, Metaxas DN, Gore JC (2011) A level set method for image segmentation in the presence of intensity inhomogeneities with application to mri. IEEE Trans Image Process 20(7):2007–2016

    MathSciNet  MATH  Article  Google Scholar 

  22. 22.

    Murofushi T, Sugeno M (2000) Fuzzy measures and fuzzy integrals. In: Fuzzy measures and integrals: theory and applications, pp 3–41

  23. 23.

    Olabarriaga SD, Smeulders AW (2001) Interaction in the segmentation of medical images: a survey. Med Image Anal 5(2):127–142

    Article  Google Scholar 

  24. 24.

    Pelekis N, Iakovidis DK, Kotsifakos EE, Kopanakis I (2008) Fuzzy clustering of intuitionistic fuzzy data. International Journal of Business Intelligence and Data Mining 3(1):45–65

    Article  Google Scholar 

  25. 25.

    Pham DL, Xu C, Prince JL (2000) Current methods in medical image segmentation 1. Annu Rev Biomed Eng 2(1):315–337

    Article  Google Scholar 

  26. 26.

    Qiu C, Xiao J, Yu L, Han L, Iqbal MN (2013) A modified interval type-2 fuzzy c-means algorithm with application in mr image segmentation. Pattern Recogn Lett 34(12):1329–1338

    Article  Google Scholar 

  27. 27.

    Sato M, Lakare S, Wan M, Kaufman A, Nakajima M (2000) A gradient magnitude based region growing algorithm for accurate segmentation. In: 2000 international conference on image processing, 2000. Proceedings, vol 3. IEEE, pp 448–451

  28. 28.

    Szmidt E, Kacprzyk J (2000) Distances between intuitionistic fuzzy sets. Fuzzy Sets Syst 114(3):505–518

    MathSciNet  MATH  Article  Google Scholar 

  29. 29.

    Verma H, Agrawal R (2015) Possibilistic intuitionistic fuzzy c-means clustering algorithm for mri brain image segmentation. Int J Artif Intell Tools 24(05):1550,016

    Article  Google Scholar 

  30. 30.

    Verma H, Agrawal R, Sharan A (2016) An improved intuitionistic fuzzy c-means clustering algorithm incorporating local information for brain image segmentation. Appl Soft Comput 46:543–557

    Article  Google Scholar 

  31. 31.

    Vlachos IK, Sergiadis GD (2007) The role of entropy in intuitionistic fuzzy contrast enhancement. In: International fuzzy systems association world congress. Springer, pp 104–113

  32. 32.

    Vovk U, Pernus F, Likar B (2007) A review of methods for correction of intensity inhomogeneity in mri. IEEE Trans Med Imaging 26(3):405–421

    Article  Google Scholar 

  33. 33.

    Wang L, Chen Y, Pan X, Hong X, Xia D (2010) Level set segmentation of brain magnetic resonance images based on local gaussian distribution fitting energy. J Neurosci Methods 188(2):316–325

    Article  Google Scholar 

  34. 34.

    Wang Z, Song Q, Soh YC, Sim K (2013) An adaptive spatial information-theoretic fuzzy clustering algorithm for image segmentation. Comput Vis Image Underst 117(10):1412–1420

    Article  Google Scholar 

  35. 35.

    Xu Z, Wu J (2010) Intuitionistic fuzzy c-means clustering algorithms. J Syst Eng Electron 21(4):580–590

    Article  Google Scholar 

  36. 36.

    Yager RR (1979) On the measure of fuzziness and negation part i: membership in the unit interval. Int J Gen Syst 5(4):221–229

    MATH  Article  Google Scholar 

  37. 37.

    Yager RR (1980) On the measure of fuzziness and negation. II. Lattices. Inf Control 44(3):236–260

    MathSciNet  MATH  Article  Google Scholar 

  38. 38.

    Zadeh LA (1965) Fuzzy sets. Inf Control 8(3):338–353

    MATH  Article  Google Scholar 

Download references

Acknowledgements

The authors would like to thank CSIR (Grant no. 09/263(1016)/2014-EMR-I) and DST PURSE for the financial support. The authors are also thankful to the anonymous reviewers for their constructive suggestions.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Dhirendra Kumar.

Ethics declarations

Conflict of interests

The authors declare that they have no conflict of interest.

Appendix A: Derivation for the membership value and cluster center

Appendix A: Derivation for the membership value and cluster center

The Lagrangian for the objective function (18) can be given as

$$\begin{array}{@{}rcl@{}} L &=& {\sum}_{i = 1}^{c} \sum\limits_{j = 1}^{N} \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m}\| {\mathbf{x}}_{j}^{IFS} - {\mathbf{v}}_{i}^{IFS}\|^{2} + {\sum}_{j = 1}^{N} Y_{j} (1 - {\sum}_{i = 1}^{c} \mu_{ij}) \end{array} $$
(43)
$$\begin{array}{@{}rcl@{}} \frac{\partial L}{\partial \mu_{ij}} &=& m \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m-1} \left[\frac{(\lambda + 1)}{1+\lambda{\mu_{ij}}} - \frac{\lambda (\lambda + 1)\mu_{ij}}{(1+\lambda{\mu_{ij}})^{2}}\right] \left( \|{\mathbf{x}}_{j}^{IFS} - {\mathbf{v}}_{i}^{IFS}\|^{2} \right) - Y_{j} = 0\\ &\Rightarrow& m \left( \frac{(\lambda + 1)}{1+\lambda{\mu_{ij}}}\right)^{m} {\mu}_{ij}^{m-1} \left[ \frac{1}{(1+\lambda{\mu_{ij}})}\right] \left( \|{\mathbf{x}}_{j}^{IFS} - {\mathbf{v}}_{i}^{IFS}\|^{2} \right) - Y_{j} = 0\\ &\Rightarrow& {\mu}_{ij}^{m-1} = \left( \frac{Y_{j}}{m(\lambda + 1)^{m}}\right) \left[\frac{(1+\lambda{\mu_{ij}})^{m + 1}}{\left( \|{\mathbf{x}}_{j}^{IFS} - {\mathbf{v}}_{i}^{IFS}\|^{2} \right)}\right]\\ &\Rightarrow& \mu_{ij} = \left( \frac{Y_{j}}{m(\lambda + 1)^{m}}\right)^{\frac{1}{m-1}} \left[\frac{(1+\lambda{\mu_{ij}})^{\frac{m + 1}{m-1}}}{\left( \|{\mathbf{x}}_{j}^{IFS} - {\mathbf{v}}_{i}^{IFS}\|^{2} \right)^{\frac{1}{m-1}}}\right]\\ &\Rightarrow& {\sum}_{i = 1}^{c} \mu_{ij} = {\sum}_{i = 1}^{c} \left( \frac{Y_{j}}{m(\lambda + 1)^{m}}\right)^{\frac{1}{m-1}} \left[\frac{(1+\lambda{\mu_{ij}})^{\frac{m + 1}{m-1}}}{\left( \|{\mathbf{x}}_{j}^{IFS} - {\mathbf{v}}_{i}^{IFS}\|^{2} \right)^{\frac{1}{m-1}}}\right]= 1\end{array} $$
$$\begin{array}{@{}rcl@{}} &\Rightarrow& \left( \frac{Y_{j}}{m(\lambda + 1)^{m}}\right)^{\frac{1}{m-1}} = \frac{1}{\sum\limits_{i = 1}^{c} \frac{(1+\lambda{\mu_{ij}})^{\frac{m + 1}{m-1}}}{\left( \|{\mathbf{x}}_{j}^{IFS} - {\mathbf{v}}_{i}^{IFS}\|^{2} \right)^{\frac{1}{m-1}}}}\\ &\Rightarrow& \mu_{ij} = \frac{ \frac{(1+\lambda{\mu_{ij}})^{\frac{m + 1}{m-1}}}{\left( \|{\mathbf{x}}_{j}^{IFS} - {\mathbf{v}}_{i}^{IFS}\|^{2} \right)^{\frac{1}{m-1}}}}{\sum\limits_{l = 1}^{c} \frac{(1+\lambda{u_{lj}})^{\frac{m + 1}{m-1}}}{\left( \|{\mathbf{x}}_{j}^{IFS} - \mathbf{v}^{IFS}_{l}\|^{2}\right)^{\frac{1}{m-1}}}}\\ &\Rightarrow& \mu_{ij} = \frac{1}{\sum\limits_{l = 1}^{c} \left( \frac{\|{\mathbf{x}}_{j}^{IFS} - {{\mathbf{v}}_{i}^{IFS}}\|}{\|\mathbf{x}_{j}^{IFS} - {\mathbf{v}^{IFS}_{l}}\|}\right)^{\frac{2}{m-1}} \left( \frac{1 + \lambda u_{lj}}{1 + \lambda \mu_{ij}}\right)^{\frac{m + 1}{m-1}}} \end{array} $$
(44)
$$\begin{array}{@{}rcl@{}} \frac{\partial L}{\partial \mu_{V}(v_{i})} &=& {\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m} (\mu_{X}(x_{j})-\mu_{V}(v_{i}))= 0 \end{array} $$
(45)
$$\begin{array}{@{}rcl@{}} &\Rightarrow& {\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m} \mu_{X}(x_{j}) - {\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1) \mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m}\mu_{V}(v_{i})= 0\\ &\Rightarrow& \mu_{V}(v_{i}) = \frac{{\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1) \mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m} \mu_{X}(x_{j})}{{\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m}} \end{array} $$
(46)
$$\begin{array}{@{}rcl@{}} \frac{\partial L}{\partial \nu_{V}(v_{i})} &=& {\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m} (\nu_{X}(x_{j})-\nu_{V}(v_{i}))= 0 \end{array} $$
(47)
$$\begin{array}{@{}rcl@{}} &\Rightarrow& {\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m} \nu_{X}(x_{j}) - {\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1) \mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m}\nu_{V}(v_{i})= 0\\ &\Rightarrow& \nu_{V}(v_{i}) = \frac{{\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m} \nu_{X}(x_{j})}{{\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1) \mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m}} \end{array} $$
(48)
$$\begin{array}{@{}rcl@{}} \frac{\partial L}{\partial \pi_{V}(v_{i})} &=& {\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m} (\pi_{X}(x_{j})-\pi_{V}(v_{i}))= 0 \end{array} $$
(49)
$$\begin{array}{@{}rcl@{}} &\Rightarrow& {\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m} \pi_{X}(x_{j}) - {\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1) \mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m}\pi_{V}(v_{i})= 0\\ &\Rightarrow& \pi_{V}(v_{i}) = \frac{{\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m} \pi_{X}(x_{j})}{{\sum}_{j = 1}^{N} \left( \frac{(\lambda + 1)\mu_{ij}}{1+\lambda{\mu_{ij}}}\right)^{m}} \end{array} $$
(50)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Kumar, D., Verma, H., Mehra, A. et al. A modified intuitionistic fuzzy c-means clustering approach to segment human brain MRI image. Multimed Tools Appl 78, 12663–12687 (2019). https://doi.org/10.1007/s11042-018-5954-0

Download citation

Keywords

  • Intuitionistic fuzzy sets
  • Fuzzy c-means
  • Intuitionistic fuzzy c-means
  • Hesitation degree
  • Image segmentation
  • Magnetic resonance imaging