Abdulhamit SUBAŞI, Ahmet ALKAN, Etem KÖKLÜKAYA

EEG işaretlerinin dalgacık sinir ağı ile sınıflandırılması

Epilepsi terapisinde EEG en önemli bilgi kaynağı olduğu için, birçok araştırmacı EEG'den bu amaca uygun bilgi elde etmeye çalışmışlardır. Bu çalışmada sunulan yeni metotta, EEG'de en yüksek olabilirlik kestirimi(maximum likelihood estimation:MLE) kullanılarak elde edilen öz bağlanımlı (AR) ve hızlı fourier dönüşümü (HFD) metotları uygulanarak katsayılar elde edilmiştir. Bu katsayılar dalgacık sinir ağına girilerek, çıkışta epileptik veya değil şeklinde sınıflandırma gerçekleştirilmiştir. İşaretteki özelliklerin belirlenerek hekime tanı işleminde yardımcı olacak, otomatik bir sistem elde edilmesi amaçlanmıştır. MLE kullanan öz bağlanımlı yöntem ile dalgacık sinir ağı kullanılarak, yeni ve güvenli bir sınıflandırıcı mimarisi elde edilmiştir. Ağ, morlet ana dalgacığı temel fonksiyonunu düğüm aktivasyon fonksiyonu olarak kullanan, geri yayılımlı sinir ağıyla oluşturulmuştur. Dalgacık sinir ağıyla yapılan EEG sınıflandırmasının daha iyi sonuçlar verdiği ve bu sonuçların hastalık teşhisinde kullanılabileceği görülmüştür.

Wavelet neural network classification of EEG signals

Since EEG is one of the most important sources of information in therapy of epilepsy, several researchers tried to address the issue of decision support for such a data. We present a novel method for classifying epilepsy of full spectrum EEG recordings. This novel method uses autoregressive (AR) model by using maximum likelihood estimation (MLE) of EEG signals as an input to a wavelet neural networks (WNNs) with two discrete outputs: epileptic seizure or non-epileptic seizure. By identifying features in the signal we want to provide an automatic system that will support a physician in the diagnosing process. By applying AR with MLE in connection with wavelet neural network, a novel and reliable classifier architecture is obtained. The network is constructed by the error backpropagation neural network using Morlet mother wavelet basic function as node activation function. It is observed that, WNN classification of EEG signals gives better results and these results can also be used for diagnosis of diseases.

PDF

___

1.Guler, I., Kiymik, M. K., Akin, M. & Alkan, A. AR spectral analysis of EEG signals by using maximum likelihood estimation, Computers in Biology and Medicine, 2001, 31, 441-450.

2.Zoubir, M., & Boashash, B. Seizure Detection of Newborn EEG Using a Model Approach. IEEE Transactions on Biomedical Engineering, 1998, 45, 673-685.

3.Weng, W. & Khorasani, K. An adaptive structure neural network with application to EEG automatic seizure detection. Neural Networks, 1996, 9, 1223-1240.

4.Pradhan, N., Sadasivan, P. K. & Arunodaya,G. R. Detection of seizure activity in EEG by an artificial neural network: A preliminary study, Computers and Biomedical Research, 1996,29, 303-313.

5.Gotman, J. Automatic recognition of epileptic seizures in the EEG. Electroencephalography and Clinical Neurophysiology, 1982,54, 530-540.

6.Qu , H. & Gotman, J. A Patient-Specific Algorithm for the Detection of Seizure Onset in Long-Term EEG Monitoring: Possible Use as a Warning Device. IEEE Transactions on Biomedical Engineering, 1997,44,115-122.

7.Petrosian, A., Prokhorov, D., Homan, R., Dashei, R., & Wunsch, D. Recurrent neural network based prediction of epileptic seizures in intra- and extracranial EEG. Neurocomputing, 2000,30, 201-218.

8.Kalayci, T. & Ozdamar, O. Wavelet preprocessing for automated neural network detection of EEG spikes., IEEE Engineering in Medicine and Biology Magazine 1995,16, 0-166.

9.Instrumentation Reference and Catalogue, 1997. National Instruments .

10.Pati, Y.C. & Krishnaparasad, P.S. Analysis and synthesis of feedforward neural networks using discrete affine wavelet transformations, IEEE Transactions Neural Networks, 1993,4 (1), 73-85.

11.Zhang, Q. Using wavelet network in nonparametric estimation, IEEE Trans. Neural Networks, 1997 ,8 (2), 227-236.

12.Xu, J. & Ho, D. W.C. A basis selection algorithm for wavelet neural Networks, Neurocomputing, 2002, 48, 681-689

13.Wong, K. & Leung, A. On-line successive synthesis of wavelet Networks, Neural Processing Letters, 1998,7, 91-100.

14.Zhang, Q. & Benveniste, A. Wavelet Networks, IEEE Trans. Neural Networks, 1992, 3 (6), 889-898.

15.Sureshbabu, A. & Farrell, J.A.Wavelet-based system identification for nonlinear control IEEE Trans. Autom. Control, 1999, (2) 412-417.

16.Zhang, J., Walter, G.G. & Lee, W. Wavelet neural networks for function learning IEEE Trans. Signal Process, 43 (6), 1995, 1485-1497.

17.Qussar, Y. & Dreyfus, G. Initialization by selection for wavelet network training, Neurocomputing, 2000, 34, 131-143

18.Daubechies, I. Ten Lectures on Wavelets. CBMS-NSF Regional Series in Applied Mathematics, SIAM, 1992, Philadelphia, PA, USA.