EŞİT GENİŞLİKLİ AYRIKLAŞTIRMA YÖNTEMİNE DAYALI YENİ BİR ÖZELLİK ÇIKARTMA YAKLAŞIMI VE YAPAY SİNİR AĞI KULLANARAK EPİLEPTİK ATAK TESPİTİ

Öz

Bu çalışmada, eşit genişlikli ayrıklaştırma (EGA) yöntemine dayalı yeni bir özellik çıkartma yaklaşımı önerilmişve Elektroensefalogram (EEG) işaretlerinden epileptik atak tespitinde bu yaklaşım ile elde edilen istatikselözellikler çok katmanlı algılayıcı sinir ağı (ÇKASA) modeline giriş olarak kullanılmıştır. Bu amaç için, EEGişaretleri EGA yöntemi ile ayrıklaştırılmış, her ayrık bölgenin yoğunluğuna dayalı histogramları elde edilmiş vehem gizli katmansız hem de 5 nörona sahip tek gizli katmanlı iki ÇKASA modeline giriş vektörü olarakuygulanmıştır. Her iki ÇKASA modeli de epileptik atak tespitinde yüksek başarı sağlamıştır. Bu sonuç, önerilenözellik çıkartma yöntemi sayesinde, doğrusal sınıflayıcıların da epileptik atak tespiti problemini çözebileceğinigöstermektedir. Sonuç olarak, EGA histogramı yaklaşımı biyomedikal işaret işlemede yeni bir özellik çıkartmayöntemi olarak kullanılabilir.

Anahtar Kelimeler

• Eşit genişlikli ayrıklaştırma, biyomedikal işaret işleme, EEG işaretleri, epileptik atak tespiti, çok katmanlı algılayıcı sinir ağı, histogram

Kaynakça

1. Adeli, H., Zhou, Z., ve Dadmehr, N., “Analysis
2. of EEE Records in an Epileptic Patient using Wavelet Transform”, Journal of Neuroscience
3. Methods, No 123, 69–87, 2003.
4. Güler, İ., ve Übeyli, E. D., “Adaptive Neuro-
5. Fuzzy Inference System for Classification of
6. EEG Signals using Wavelet Coefficients”,
7. Journal of Neuroscience Methods, No 148,
8. –121, 2005.

9. Khan, Y. U., ve Gotman, J., “Wavelet Based
10. Automatic Seizure Detection in Intra-cerebral
11. Electroencephalogram”, Clinical Neurophysiology,
12. No 114, 898–908, 2003.
13. Kıymık, M. K., Akın, M., ve Subaşı, A.,
14. “Automatic Recognition of Alertness Level by
15. using Wavelet Transform and Artificial Neural
16. Network”, Journal of Neuroscience Methods,
17. No 139, 231–240, 2004.
18. Ocak, H., “Automatic Detection of Epileptic
19. Seizures in EEG using Discrete Wavelet
20. Transform and Approximate Entropy”, Expert
21. Systems with Applications, No 36, 2027–2036,
22. -
23. Subaşı, A., “Automatic Recognition of Alertness
24. Level from EEG by using Neural Network and
25. Wavelet Coefficients”, Expert Systems with
26. Applications, No 28, 701–711, 2005.
27. Subaşı, A., “Epileptic Seizure Detection using
28. Dynamic Wavelet Network”, Expert Systems
29. with Applications, No 29, 343–355, 2005.
30. Subaşı, A., “Automatic Detection of Epileptic
31. Seizure using Dynamic Fuzzy Neural Networks”,
32. Expert Systems with Applications, No 31, 320–
33. , 2006.
34. Subaşı, A., “EEG Signal Classification using
35. Wavelet Feature Extraction and a Mixture of
36. Expert Model”, Expert Systems with
37. Applications, No 32, 1084–1093, 2007.
38. Übeyli, E. D., “Combined Neural Network Model
39. Employing Wavelet Coefficients for EEG Signals
40. Classification”, Digital Signal Processing, No
41. , 297–308, 2009.
42. Übeyli, E. D., “Decision Support Systems for
43. Time-Varying Biomedical Signals: EEG Signals
44. Classification”, Expert Systems with
45. Applications, No 36, 2275–2284, 2009.
46. Alkan, A., Köklükaya, E., ve Subaşı, A.,
47. “Automatic Seizure Detection in EEG using
48. Logistic Regression and Artificial Neural
49. Network”, Journal of Neuroscience Methods,
50. No 148, 167–176, 2005.
51. Subaşı, A., ve Ercelebi, E., “Classification of
52. EEG Signals using Neural Network and Logistic
53. Regression”, Computer Methods and
54. Programs in Biomedicine, No 78, 87–99, 2005.
55. Altunay, S., Telatar, Z., ve Eroğul, O., “Epileptic
56. EEG Detection using the Linear Prediction Error
57. Energy”, Expert Systems with Applications,
58. Cilt 37, No 8, 5661–5665, 2010.
59. Acır, N., “Automated System for Detection of
60. Epileptiform Patterns in EEG by using a
61. Modified RBFN Classifier”, Expert Systems
62. with Applications, Cilt 29, No 2, 455–462, 2005.
63. Subaşı, A., “Application of Adaptive Neuro-
64. Fuzzy Inference System for Epileptic Seizure
65. Detection using Wavelet Feature Extraction”,
66. Computers in Biology and Medicine, Cilt 37,
67. No 2, 227–244, 2007.
68. Aslan, K., Bozdemir, H., Şahin, S., Oğulata, S. N.,
69. ve Erol, R., “A Radial Basis Function Neural
70. Network Model for Classification of Epilepsy
71. using EEG Signals”, The Journal of Medical
72. Systems, No 32, 403–408, 2008.
73. Petrosian, A., Prokhorov, D., Homan, R., Dashei,
74. R., ve Wunsch, D., “Recurrent Neural Network
75. Based Prediction of Epileptic Seizures in Intra
76. and Extra-cranial EEG”, Neurocomputing, No
77. , 201–218, 2000.
78. Srinivasan, V., Eswaran, C., ve Sriraam, N.,
79. “Artificial Neural Network based Epileptic
80. Detection using Time-Domain and Frequency-
81. Domain Features”, Journal of Medical Systems,
82. Cilt 29, No 6, 647–660, 2005.
83. Pradhan, N., Sadasivan, P. K., ve Arunodaya, G.
84. R., “Detection of Seizure Activity in EEG by an
85. Artificial Neural Network: A Preliminary Study”,
86. Computers and Biomedical Research, No 29,
87. –313, 1996.
88. Übeyli, E. D., “Analysis of EEG Signals by
89. Combining Eigenvector Methods and Multi-class
90. Support Vector Machines”, Computers in
91. Biology and Medicine, Cilt 38, No 1, 14–22,
92. -
93. Kıymık, M. K., Subaşı, A., ve Özçalık, H. R.,
94. “Neural Networks with Periodogram and
95. Autoregressive Spectral Analysis Methods in
96. Detection of Epileptic Seizure”, Journal of
97. Medical Systems, Cilt 28, No 6, 511–522, 2004.
98. Kıymık, M. K., Güler, İ., Dizibüyük, A., and
99. Akın, M., “Comparison of STFT and Wavelet
100. Transform Methods in Determining Epileptic
101. Seizure Activity in EEG Signals for Real-time
102. Application”, Computers in Biology and
103. Medicine, Cilt 35, No 7, 603-616, 2005.
104. Andrzejak, R. G., Lehnertz, K., Mormann, F.,
105. Rieke, C., David, P., ve Elger, C. E., “Indications
106. of Nonlinear Deterministic and Finite-
107. Dimensional Structures in Time Series of Brain
108. Electrical Activity: Dependence on Recording
109. Region and Brain State”, Physical Review E,
110. Cilt 64, No 6, 061907, 2001.
111. Hsu, C. N., Huang, H., J., Wong, T. T.,
112. “Implications of the Dirichlet Assumption for
113. Discretization of Continuous Variables in Naive
114. Bayesian Classifiers”, Machine Learning, Cilt
115. , No 3, 235–263, 2003.
116. Jiang, S., Li, X., Zheng, Q., ve Wang, L.,
117. “Approximate Equal Frequency Discretization
118. Method”, Global Congress on Intelligent
119. Systems, 514-518, 2009.
120. Tay, E. H., ve Shen, L., “A Modified Chi2
121. Algorithm for Discretization”, IEEE
122. Transactions on Knowledge and Data
123. Engineering, Cilt 14, No 3, 666–670, 2002.
124. Gang, L. ve Tong, F., “An Unsupervised
125. Discretization Algorithm based on Mixture
126. Probabilistic Model”, Chinese Journal of
127. Computers, Cilt 25, No 2, 158–164, 2002.
128. Lee, C.H., “A Hellinger-based Discretization
129. Method for Numeric Attributes in Classification
130. Learning”, Knowledge-Based Systems, Cilt 20,
131. No 4, 419–425, 2007.
132. Fayyad, U.M, ve Irani, K.B., “Multi-interval
133. Discretization of Continuous Valued Attributes
134. for Classification Learning”, Proc. of the 13th
135. International Joint Conference on Artificial
136. Intelligence, 1022–1029, 1993.
137. Clarke, E.J. ve Braton, B.A., “Entropy and MDL
138. Discretization of Continuous Variables for
139. Bayesian Belief Networks”, International
140. Journal of Intelligence Systems, No 15, 61–92,
141. -
142. Xi, J. Ouyang, W.M., “Clustering based
143. Algorithm for Best Discretizing Continuous
144. Valued Attributes”, Mini-micro systems, Cilt 21,
145. No 10, 1025–1027, 2000.