Abstract

Magnetic Resonance Imaging (MRI) has become an effective tool for clinical research in recent years and has found itself in applications such as brain tumour detection. In this study, tumor classification using multiple kernel-based probabilistic clustering and deep learning classifier is proposed. The proposed technique consists of three modules, namely segmentation module, feature extraction module and classification module. Initially, the MRI image is pre-processed to make it fit for segmentation and de-noising process is carried out using median filter. Then, pre-processed image is segmented using Multiple Kernel based Probabilistic Clustering (MKPC). Subsequently, features are extracted for every segment based on the shape, texture and intensity. After features extraction, important features will be selected using Linear Discriminant Analysis (LDA) for classification purpose. Finally, deep learning classifier is employed for classification into tumor or non-tumor. The proposed technique is evaluated using sensitivity, specificity and accuracy. The proposed technique results are also compared with existing technique which uses Feed-Forward Back Propagation Network (FFBN). The proposed technique achieved an average sensitivity, specificity and accuracy of 0.88, 0.80 and 0.83, respectively with the highest values as about 1, 0.85 and 0.94. Improved results show the efficiency of the proposed technique.

Keywords:

Deep learning classifier, Linear Discriminant Analysis (LDA), MRI image segmentation, Multiple Kernel based Probabilistic Clustering (MKPC), shape and texture based features , tumour detection,

References

1. Acir, N., O. Ozdamar and C. Guzelis, 2006. Automatic classification of auditory brainstem responses using SVM-based feature selection algorithm for threshold detection. Eng. Appl. Artif. Intel., 19: 209-218.
   CrossRef    
2. Ali, S.M., L.K. Abood and R.S. Abdoon, 2013. Brain tumor extraction in MRI images using clustering and morphological operations techniques. Int. J. Geogr. Inform. Syst. Appl. Remote Sens., 4(1).
   
3. Bankman, I.H., 2009. Handbook of Medical Image processing and Analysis. Academic Press in Biomedical Engineering, Burlington, MA 01803, USA.
   
4. Georgiardis, P., D. Cavouras, I. Kalatzis, A. Daskalakis, G.C. Kagadis, M. Malamas, G. Nikifordis and E. Solomou, 2008. Improving brain tumor characterization on MRI by probabilistic neural networks on non-linear transformation of textural features. Comput. Meth. Prog. Bio., 89: 24-32.
   CrossRef    PMid:18053610    
5. Gholami, B., I. Norton, L.S. Eberlin and N.Y.R. Agar, 2013. A statistical modeling approach for tumor-type identi?cation in surgical neuropathology using tissue mass spectrometry imaging. IEEE J. Biomed. Health Inform., 17(3): 734-744.
   CrossRef    PMid:24592474    
6. Karabatak, M. and M.C. Ince, 2009. An expert system for detection of breast cancer based on association rules and neural network. Expert Syst. Appl., 36: 3465-3469.
   CrossRef    
7. Kharat, K.D., P.P. Kulkarni and M.B. Nagori, 2012. Brain tumor classification using neural network based methods. Int. J. Comput. Sci. Inform., 1(4).
   
8. Sapra, P., R. Singh and S. Khurana, 2013. Brain TUMOR detection using neural network. Int. J. Sci. Modern Eng. (IJISME), 1(9).
   
9. Sharma, M., R.B. Dubey, Sujata and S.K. Gupta, 2012. Feature extraction of mammogram. Int. J. Adv. Comput. Res., 2(5).
   
10. Sridhar, D. and M. Krishna, 2013. Brain tumor classification using discrete cosine transform and probabilistic neural network. Proceeding of International Conference on Signal Processing, Image Processing and Pattern Recognition (ICSIPR, 2013), pp: 92-96.
    CrossRef    
11. Sun, Z., C. Zheng, Q. Gao, J. Zhang and D. Zhan, 2012. Tumor classi?cation using eigengene-based classi?er committee learning algorithm. IEEE Signal Process. Lett., 19(8).
    
12. Wang, S.L., Y.H. Zhu, W. Jia and D.S. Huang, 2012. Robust classification method of tumor subtype by using correlation filters. IEEE-ACM T. Comput. Bi., 9(2).
    

Competing interests

The authors have no competing interests.

Open Access Policy

This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Copyright

The authors have no competing interests.