Open Access Open Access  Restricted Access Subscription or Fee Access

Identification of Brain Functions and its Disorders Using SVM

S. Sathiya, V. Rathikarani, Dr. M. Balasubramanian, Dr. P. Dhanalakshmi

Abstract


This paper helps the neurologists to diagnose and determine weather a patient is epileptic or non epileptic .EEG standard is an acronym for Electroencephalogram has been traditionally used to diagnose the brain activity of a patient which corresponds to epilepsy and other brain disorders. The features are extracted using medical data signals and SVM is trained using the extracted signals. This paper focuses on developing a classification technique called SVM for multichannel EEG recordings. Traditional SVMs are used to exploit both spatial and temporal characteristics of EEG data. The proposed technique is tested on two EEG data sets acquired from ten patients respectively [1] SVM is based on the principle of structural risk minimization and it can be used for non linear regression. This study is a necessary application of data mining to advance the diagnosis and treatment of human epilepsy. The performance system achieves the identification rate of 96.0 % for 10 subjects.

Keywords


Electroencephalogram (EEG) Classification, Epilepsy Diagnosis, Discrete Wavelet Transform and Support Vector Machine.

Full Text:

PDF

References


Wanpracha Art Chaovalitwongse, , Rebecca S. Ya-Ju Fan, Pottenger, youyi Wang, and Leon D. Iasemidis, “Pattern- and Network-Based Classification Techniques for Multichannel Medical Data Signals to Improve Brain Diagnosis” 1083-4427/$26.00 © 2011 IEEE

W. Chaovalitwongse, P. Pardalos, and O. Prokoyev, “Electroencephalogram (EEG) time series classification: Applications in epilepsy,” Ann. Oper. Res., vol. 148, no. 1, pp. 227–250, Nov. 2006.

W. Chaovalitwongse, Y. Fan, and R. Sachdeo, “On the time series k-nearest neighbor for abnormal brain activity classification,” IEEE Trans. Syst., Man, Cybern. A, Syst., Humans, vol. 37, no. 6, pp. 1005–1016, Nov. 2007.

W. Chaovalitwongse, Y. Fan, and R. Sachdeo, “Support feature machine for classification of abnormal brain activity,” in Proc. 13th ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2007, pp. 113–122.

W. Chaovalitwongse, Y. Fan, and R. Sachdeo, “Novel optimization models for abnormal brain activity classification,” Oper. Res., vol. 56, no. 6, pp. 1450–1460, Nov./Dec. 2008.

Abdulhamit Subasi, “EEG signal classification using wavelet feature extraction and a mixture of expert model”, Department of Electrical and Electronics Engineering, Kahramanmaras Sutcu Imam University, Avsar Yerleskesi, 46050-9 Kahramanmaraş, Turkey .28 February 2006.

M. R. Nazari Kousarrizi, A. Asadi Ghanbari, K. N. Toosi, M. Teshnehlab, M. Aliyari, A. Gharaviri, “Feature Extraction and Classification of EEG Signals using Wavelet Transform, SVM and Artificial Neural Networks for Brain Computer Interfaces” , Azad University Science and Research branch , 2009

Matthias Honal and Tanja Schultz ,”DETERMINE TASK DEMAND FROM BRAIN ACTIVITY” Karlsruhe University, Am Fasanengarten 5, 78131,2008

S. Baik and J. Bala, “A decision tree algorithm for distributed data mining: Towards network intrusion detection,” in Proc. Int. Conf. Comput. Sci. Appl., 2004, vol. 3046, pp. 206–212.

N. Cristianini and J. Taylor, An Introduction to Support Vector Machines and Other Kernel-Based Learning Methods. Cambridge, U.K.: Cambridge Univ. Press, 2000.

K. P. Bennett and O. L. Mangasarian, “Robust linear programming discrimination of two linearly inseparable sets,” Optim. Methods Softw., vol. 1, no. 1, pp. 23–34, 1992.

P. Bradley, U. Fayyad, and O. Mangasarian, “Mathematical programmingfor data mining: Formulations and challenges,” INFORMS J. Comput.,vol. 11, no. 3, pp. 217–238, 1999.

Joachims, T. (1999). Making Large-Scale SVM Learning Practical, chapter 11.MIT-Press.Jung, T., Makeig, S., Humphries, C., Lee, T., Mckeown,

M., Iragui, V., Sejnowski, T. (2000) Removing Electroencephalographic Artifacts by Blind Source Separation. Psychophysiology, 37(2):163-17.


Refbacks

  • There are currently no refbacks.