Abstract

In this paper, electroencephalographic (EEG) signals are analyzed and classified based on a new multilevel transfer function quantum wavelet neural network (QWNN) model. The independent component analysis (ICA) is used as processing after normalization of these signals. Some features are extracted from the data using the clustering technique (CT). The classification result of the new model is compared with that of wavelet neural network (WNN), quantum neural network (QNN), and feed forward neural network (FFNN). The new QWNN model is found to achieve average classification accuracy of 94. 187%, but classification accuracies using WNN, QNN and FFNN are 89. 803%, 83. 713% and 75. 076%, respectively

References

• Ropper, A. H. , and Brown, R. H. 2011 Adams and Victor's Principles of Neurology. 8th ed.
• Hauser, S. L. , and Josephson, S. A. 2011 Harrison Neurology in Clinical Medicine, 2nd ed. San Francisco: University of California.
• Verma, A. K. , and Mangaraj, A. K. 2010 Analysis and Classification of Electroencephalography Signals. M. Sc. thesis, Electrics and Communication Eng. Dept. , National Institute of Technology, Rourkela.
• Adikarapatti, V. 2007 Optimal EEG Channels and Rhythm Selection for Task Classification. Madras University, India, April 2007.
• Bartosova, V. , Vysata, O. and Prochazka, A. 2006 Graphical User Interface for EEG Signal Segmentation. Computing and Control Eng. Dept. , Institute of Chemical Technology.
• Zhou, J. , Gan, Q. , Krzyzak, A. , and Suen, C. Y. 1999. Recognition of handwriting numerals by quantum neural network with fuzzy features. International Journal on Document Analysis and Recognition, ©Springer-Verlag, no. 2, 30-36.
• Karayiannis, N. B. , Mukherjee, A. , Glover, J. R. , Frost, Jr, J. D. , Hrachovy, R. A. , and Mizrahi, E. M. 2005. An evaluation of quantum neural networks in the detection of epileptic seizures in the neonatal electroencephalogram. Soft Comput. , Springer-Verlag,(10 May 2005).
• Liu, K. , Peng, L. and Yang, O. 2010. The algorithm and application of quantum wavelet neural networks. in 2010 Chinese Control and Decision Conference, pp. 2941-2945.
• Karayiannis, N. B. and Purushothaman, G. 1997 Quantum neural networks (QNN's): inherently fuzzy feed forward neural networks. IEEE Trans. Neural Networks. vol. 8, no. 3, May 1997.
• Xianwen, R. , Feng, Z. , Lingfeng, Z. and Xianwen, M. 2010. Application of quantum neural network based on rough set in transformer fault diagnosis. In Proc. Asia-Pacific Power and Engineering Conference, 2010, pp. 1-4.
• Malinowski, A. , Cholewo, T. J. and Zurada, J. M. 1995. Capabilities and limitations of feed forward neural networks with multilevel neurons. In Proc. of the IEEE International Symposium on Circuits and Systems, vol. 1, Seattle, Wash. , USA, April 1995, pp. 131-134.
• (2005, September 6). More on regression gradient descent classification (COMP-652,Lecture2)[Online]. Available: http://www. facweb. iitkgp. ernet. in/~sudeshna/courses/ML06/regression-mcgill. pdf,
• Samarasinghe, S. 2006 Neural Networks for Applied Sciences and Engineering From Fundamental to Complex Pattern Recognition. Mechanical Eng. Dept. , Lumumba Univ.
• Zhu D. and Wu, R. 2007. A multilayer quantum neural networks recognition system for handwritten digital recognition. In Proc. of Third International Conference on Natural Computation (ICNC), 2007, vol. 1, pp. 718-722.
• (2005, November). EEG time series (epileptic data). [Online]. Available:http://www. meb. Unibonn. de/epileptologie/science/physikleegdata. html
• Horlings, R. 2008 Emotion Recognition Using Brain Activity. Faculty Electrical Engineering, Mathematics, and Computer Science, Man-machine Interaction Group, Delft University of Technology, March 2008.
• Principal component analysis (PCA) and independent component analysis (ICA). [Online]. Available:http://www. cis. hut. fi/projects/ica/fastica/
• Faul, S. D. 2007 Automated Neonatal Seizure Detection. M. Sc. thesis, Electrical and Electronic Eng. Dept. , National University of Ireland, Cork, 1st August 2007.
• Feature extraction and selection methods. [Online]. Available:http://www. isa. umh. es/asignaturas/cscs/PR/3%20-%20Feature%20extraction. pdf
• Siuly L. Y. and Wen, P. 2010. Analysis and classification of EEG signals using a hybrid cluster technique. In Proc. IEE/ICME International Conference on Complex Medical Engineering, Gold Cost, Australia, July 13-15, pp. 34-39.