An Improved k Nearest Neighbor Classifier Using Interestingness Measures for Medical Image Mining

Authors: J. Alamelu Mangai, Satej Wagle, V. Santhosh Kumar

Abstract:

The exponential increase in the volume of medical image database has imposed new challenges to clinical routine in maintaining patient history, diagnosis, treatment and monitoring. With the advent of data mining and machine learning techniques it is possible to automate and/or assist physicians in clinical diagnosis. In this research a medical image classification framework using data mining techniques is proposed. It involves feature extraction, feature selection, feature discretization and classification. In the classification phase, the performance of the traditional kNN k nearest neighbor classifier is improved using a feature weighting scheme and a distance weighted voting instead of simple majority voting. Feature weights are calculated using the interestingness measures used in association rule mining. Experiments on the retinal fundus images show that the proposed framework improves the classification accuracy of traditional kNN from 78.57 % to 92.85 %.

Keywords: Medical Image Mining, Data Mining, Feature Weighting, Association Rule Mining, k nearest neighbor classifier.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1087652

References:

[1] W. Hsu, M. L. Lee, and J. Zhang, “Image Mining : Trends & Developments.” in J of Intellident Information Systems, vol 19, issue 1, , pp. 7-23, 2002..
[2] A. S. Elmaghraby, M. M. Kantardzic, and M P. Wachowiak, “Data Mining from Multimedia Patient Records”, Data Mining and Knowledge Discovery Approaches based on Rule Induction Techniques, Springer, pp. 551–595, 2006.
[3] M.-L. Antonie, O. Zaiane, and A. Coman, “Application of Data Mining Techniques for Medical Image Classification”, in Proc of 2nd Intl. Workshop on Multimedia Data Mining, 2001, pp. 94-101.
[4] M. Vasantha, V. S. Bharathi and R. Dhamodaran, “Medical Image Feature Extraction, Selection and Classification”, In. J of Engineering Science and Technologl, vol. 2, no. 6, pp. 2071 – 2076, 2010
[5] P. Rajendran, and M. Madheswaran, “Hybrid Medical Image Classification Using Association Rule Mining with Decision Tree Algorithm”, J. of Computing, vol 2, no 1, pp. 127–136, 2010.
[6] A. Kharrat, K. Gasmi, M. B. Messaoud, N. Benamrane and M. Abid, “A Hybrid Approach for Automatic Classification of Brain MRI Using Genetic Algorithm and Support Vector Machine”, J. of Sciences, issue 17, pp. 71-82, 2010.
[7] B.G..Prasad and A. N. Krishna, “Classification of Medical Images using Data Mining Techniques”, Advances in Communication, Network & Computing”, Advances in Communication, Network & Computing, Lecture Notes of the Institute of Computer sciences, social informatics and Telecommunication Engineering, vol.108,, pp. 54-59, 2012.
[8] Z. S. Zubi, and R. A. Saad, “Using some Data Mining Techniques for Early Diagnosis of Lung Cancer”, in Proc. Of 10th WSEAS Intl. conference on Artificial Intelligence, Knowledge Engineering and Data Bases, 2011, pp. 32-37.
[9] N. H. Rajini, and R.Bhavani, “Classification of MRI Brain Images using k nearest neighbor and artificial neural network” IEEE-Intl conference on Recent Trends in Information Technology, pp. 863-868, 2011.
[10] L. Muflikhah, and Adnyana, “Classifying Categorical Data Using Modified K-Nearest Neighbor Weighted by Association Rules”, in Proc of the Int Conf on Future Information Technology, vol 13, 2011, pp. 347-351.
[11] J. Gou , L. Du , Y. Zhang and T. Xiong, “New Distance-weighted knearest Neighbor Classifier”, J. of Informational and Computational Science, vol 9, no 6, pp. 1426-1429, 2012.
[12] R. C. Ganzalez, amd R. E. Woods, Digital Image Processing, 2nd ed, New Jersey: Prentice Hall 2001
[13] M. Hall, E. Frank, G. Holmes, B. Pfahringer, P. Reutemann, I. H. Witten. “The WEKA data mining software: an update”, ACM SIGKDD Explorations, vol. 11, issue 1, pp. 10-18, June 2009.