A New Classification Approach Based On Support Vector Regression For Epileptic Seizure Detection

Year 2024, , 587 - 601, 27.03.2024

Esra Betül Kınacı , Hasan Bal , Harun Kınacı

https://doi.org/10.2339/politeknik.1055549

Abstract

Although the classification problem is a subject that has been studied by researchers for a long time, it is still up-to-date. Especially the problems that image processing and diagnosis of disease are some of the most current application topics. This study presents a new data classification method based on support vector regression and mathematical programming. The proposed method consists of a two-stage hybrid structure. In the first step, the classification score is obtained for each unit with the support vector regression equation. In the second stage, using the classification scores of the units, a classification rule is created with the help of a mathematical model and the classification of the units is provided. The proposed method offers an alternative innovation to traditional methods. Methods based on traditional mathematical programming separate classes with a linear function. This situation limits the use of algorithms based on mathematical programming. The proposed method can be used in all linear or non-linearly separable data structures, as well as easily transforming into problem types with more than two groups. The model is applied to the classification problem of Electroencephalograph (EEG) signals and the classification performance is compared with the existing methods. The results obtained are given in the tables and it is shown that the proposed model can be an alternative to the existing algorithms

Keywords

Support Vector Machines, Support Vector Regression, Classification, Bonn Dataset

Epileptik Nöbet Tespiti İçin Destek Regresyon Temelli Yeni Bir Sınıflandırma Yaklaşımı

Abstract

Sınıflandırma problemi araştırmacılar tarafından uzun zamandır incelenen bir konu olmasına rağmen güncelliğini hala korumaktadır. Özellikle görüntü işleme ve hastalık tanısının belirlenmesi problemleri güncel uygulama alanlarından bazılardır. Bu çalışma destek vektör regresyon ve matematiksel programlamaya dayalı yeni bir veri sınıflandırma yöntemi sunmaktadır. Önerilen yöntem iki aşamalı hibrit bir yapıdan oluşmaktadır. İlk aşamada, destek vektör regresyon denklemi ile her bir birim için sınıflandırma skoru elde edilirken ikinci aşamada ise birimlerin sınıflandırma skorları kullanılarak bir matematiksel model yardımıyla sınıflandırma kuralı oluşturulur ve birimlerin sınıflandırılması sağlanır. Önerilen yöntem geleneksel yöntemlere alternatif bir yenilik sunmaktadır. Geleneksel matematiksel programlamaya dayalı yöntemler sınıfları doğrusal bir fonksiyon ile ayırır. Bu durum ise matematiksel programlamaya dayalı algoritmalarının kullanımını kısıtlar. Önerilen yöntem, doğrusal veya doğrusal ayrılamayan veri yapılarının tamamında kullanılabilir olmasının yanı sıra ikiden fazla grup sayısının olduğu problem türlerine de kolaylıkla dönüştürülebilmektedir. Model önce simülasyon ile irdelenmiş sonrasında Elektroensefalograf (EEG) sinyallerinin sınıflandırılması probleminde uygulanmış ve sınıflandırma performansı mevcut yöntemlerle karşılaştırılmıştır. Elde edilen sonuçlar tablolarda verilmiş ve önerilen modelin mevcut algoritmalara alternatif olabileceğini gösterilmiştir.

Keywords

Destek Vektör Makineleri, Destek Vektör Regresyon, Sınıflandırma, Bonn Veri Seti

References

• [1] T.S. Kumar, V. Kanhangad, R.B. Pachori, "Classification of seizure and seizure-free EEG signals using local binary patterns", Biomedical Signal Processing and Control, 15: 33–40, (2015).
• [2] Y. Kaya, T. Ramazan, "Epileptik EEG işaretlerinin sınıflandırılması için yeni bir öznitelik çıkarım yöntemi", Süleyman Demirel Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 22:(2018).
• [3] J. Jing, X. Pang, Z. Pan, F. Fan, Z. Meng, "Classification and identification of epileptic EEG signals based on signal enhancement", Biomedical Signal Processing and Control, 71:(2022).
• [4] X. Ma, Z. Zhuo, L. Wei, Z. Ma, Z. Li, H. Li, "Altered temporal organization of brief spontaneous brain activities in patients with alzheimer’s Disease", Neuroscience, 425: 1–11, (2020).
• [5] Y. Fu, X. Xiong, C. Jiang, B. Xu, Y. Li, H. Li, "Imagined hand clenching force and speed modulate brain activity and are classified by NIRS combined with EEG", IEEE Transactions on Neural Systems and Rehabilitstion Engineering, 25: 1641–1652, (2017).
• [6] R. W. Peng, "Classification method of EEG based on emprical mode decomposition and SVM", Comp. Meas. Cont., 28: 189–194, (2020).
• [7] W. Klonowski, "Everything you wanted to ask about EEG but were afraid to get the right answer", Nonlinear Biomedical Physics, 3: 1–6, (2009).
• [8] M. Heyden, "Classification of EEG data using machine learning techniques", Lund University Press, (2016).
• [9] Ö. Türk, M.S. Özerdem, "Epileptik EEG sinyallerinin sınıflandırılması için bir boyutlu medyan yerel ikili örüntü temelli öznitelik çıkarımı", GU Journal of Sciences, Part C., 5: 97–107, (2017).
• [10] E. Tuncer, E.D. Bolat, "Destek vektör makineleri ile EEG sinyallerinden epileptik nöbet sınıflandırması Epileptic seizure classification from eeg signals with support vector machines", Politeknik Dergisi, 25(1): 239–249, (2022).
• [11] S. Ramakrishnan, A.S. Muthanantha Murugavel, P. Sathiyamurthi, J. Ramprasath, "Seizure detection with local binary pattern and CNN classifier", Journal of Physsics: Conference Series, 1767: (2021).
• [12] V. Srinivasan, C. Eswaran, A.N. Sriraam, "Artificial neural network based epileptic detection using time-domain and frequency-domain features", Journal of Medical Systems, 29: 647–660, (2005).
• [13] E.D. Übeyli, "Decision support systems for time-varying biomedical signals: EEG signals classification", Expert Systems withApplications. 36(2): 2275–2284, (2009).
• [14] H. Ocak, "Automatic detection of epileptic seizures in EEG using discrete wavelet transform and approximate entropy", Expert Systems withApplications. 36(2):2027–2036, (2009).
• [15] Y. Li, P.P. Wen, "Clustering technique based least square support vector machine for EEG signal classification", Computer Methods and Programs in Biomedicine. 104: 358–372, (2011).
• [16] S. Chandaka, A. Chatterjee, S. Munshi, "Cross-correlation aided support vector machine classifier for classification of EEG signals", Expert Systems withApplications, 36(2): 1329–1336, (2009).
• [17] I. Guler, E.D. Ubeyli, "Multiclass support vector machines for EEG-signals classification", IEEE Transactions on Information Technology Biomedicine, 11(2): 117–126, (2007).
• [18] L. Guo, D. Rivero, J. Dorada, J.R. Rabunal, A. "Pazos, Automatic epileptic seizure detection in EEGs based on line length feature and artificial neural networks", Journal of Neuroscience Methods, 191(1): 101–109, (2010).
• [19] L. Guo, D. Rivero, A. Pazos, "Epileptic seizure detection using multiwavelet transform based approximate entropy and artificial neural networks", Journal of Neuroscience Methods, 193(1): 156–163, (2010).
• [20] U. Orhan, M. Hekim, M. Özer, "Epileptic seizure detection using artificial neural network and a new feature extraction approach based on equal width discretization", Journal of Faculty Engineering and Architecture of Gazi University, 26(3): 575–580, (2011).
• [21] U. Orhan, M. Hekim, M. Ozer, "Epileptic seizure detection using probability distribution based on equal frequency discretization", Journal of. Medical Systems, 36(4): 2219–2224, (2012).
• [22] M. Hekim, "The classification of EEG signals using discretization-based entropy and the adaptive neuro-fuzzy inference system", Turkish Journal of Electrical Engineering & Computer Sciences 24: 285–297, (2016).
• [23] N. Nicolaou, J. Georgiou, "Detection of epileptic electroencephalogram based on permutation entropy and support vector machines", Expert Systems with Applications, 39: 202–209, (2012).
• [24] T.K. Gandhi, P. Chakraborty, G.G. Roy, B.K. Panigrahi, "Discrete harmony search based expert model for epileptic seizure detection in electroencephalography", Expert Systems with Applications, 39: 4055–4062, (2012).
• [25] D. Wang, D. Miao, C. Xie, "Best basis-based wavelet packet entropy feature extraction and hierarchical EEG classification for epileptic detection", Expert Systems with Applications, 38: 14314–14320, (2011).
• [26] K. Polat, S. Güneş, "Classification of epileptiform EEG using a hybrid system based on decision tree classifier and fast Fourier transform", Applied Mathematics and Computation, 18: 1017–1026, (2007).
• [27] M. Mursalin, Y. Zhang, Y. Chen, N. V. Chawla, "Automated epileptic seizure detection using improved correlation-based feature selection with random forest classifier", Neurocomputing. 241: 204–214, (2017).
• [28] S.C. Satapathy, J.V.R. Murthy, P.V.G.D. Prasad Reddy, B.B. Misra, P.K. Dash, G. Panda, "Particle swarm optimized multiple regression linear model for data classification", Applied Soft Computing, 9(2): 470–476, (2009).
• [29] M.I. Doǧan, A. Orman, M. Örkcü, H.H. Örkcü, "A new approach based on regression analysis and mathematical programming to multigroup classification problems", Journal of the Faculty of Engineering and Architecture of Gazi University, 34(4): 1939–1955, (2019).
• [30] K.. Lam, J.. Moy, "An experimental comparison of some linear programming approaches to the discriminant problem", Computers & Operations Research, 24(7): 593–599, (1997).
• [31] A. Tekerek, "Support vector machine based spam SMS detection", Politeknik Dergisi, 22(2): 779–784, (2019).
• [32] U. Köse, "Zeki optimizasyon tabanlı destek vektör makineleri ile diyabet teşhisi", Politeknik Dergisi, 22(3): 557–566, (2019).
• [33] B.E. Boser, I.M. Guyon, V.N. Vapnik, "A Training Algorithm for Optimal Margin Classifiers", Proceedings Annual Conferences of Computational Learning Theory, 144–152, (1992).
• [34] V. Cherkassky, F. Mulier, "Learning From Data: Concepts, Theory and Methods", John Wiley, New York, (1998).
• [35] V. Cherkassky, X. Shao, F. Mulier, V. Vapnik, "Model Complexity control for regression using VC generalization bounds", IEEE Transactions Neural Networks, 10: 1075–1089, (1999).
• [36] S. Haykin, "Neural networks: A comprehensive foundation", 2nd ed., Prentice Hall, Upper Saddle River, New Jersey, (2001).
• [37] S. Haykin, "Neural networks: A comprehensive foundation", 4th ed, Pearson Education, Singapore, (2003).
• [38] S. Tripathi, V. V. Srinivas, R.S. Nanjundiah, "Downscaling of precipitation for climate change scenarios: A support vector machine approach", Journal of Hydrology, 330(3-4): 621–640, (2006).
• [39] S.R. Gunn, "Support Vector Machines for Classification and Regression", (1998)
• [40] S. Haykin, "Neural networks and learning machine", 3rd ed., Pearson Prentice-Hall, New York.
• [41] A.J. Smola, B. Schölkopf, "A tutorial on support vector regression", Statistics and Computing, 14: 199–222, (2004).
• [42] J. Mercer, "Functions of positive and negative type and their connection with the theory of integral equations", Philosoptical Transactions of the Royal Society of London, 209: 415–446, (1909). [43] R. Courant, D. Hilbert, "Methods of Mathematical Physics", Wiley Interscience, New York, (1970).
• [44] N. Freed, F. Glover, "A linear programming approach to the discriminant problem", Decision Sciences, 12(1): 68–74, (1981).
• [45] N. Freed, F. Glover, "Simple but powerful goal programming models for discriminant problems", European Journal of Operational Research, 44–66, (1981).
• [46] K. Lam, E.. Choo, W. Moy, J, "Minimizing deviations from the group mean: A new linear programming approach for the two-group classification problem", European Journal of Operational Research, 358–367, (1996).
• [47] K.. Lam, J.. Moy, "Improved linear programming formulations for the multigroup discriminant problem", The Journal of the Operational Research Society, 47(12):1526–1529, (1996).
• [48] H. Bal, H.H. Örkcü, S. Çelebioǧlu, "An alternative model to Fisher and linear programming approaches in two-group classification problem: minimizing deviations from the group median", GU Journal of Sciences, 49–55, (2006).
• [49] T. Sueyoshi, "DEA-discriminant analysis in the view of goal programming", European Journal of Operational Research, 115: 564–582, (1999).
• [50] A. Stam, E.A. Joachimsthaler, "A comparison of a robust mixed-integer approach to existing methods for establishing classification rules for the discriminant problem", European Journal of Operational Research, 46: 113–122, (1990).
• [51] R. Pavur, C. Loucopoulos, "Examining optimal criterion weights in mixed integer programming approaches to the multi group classification problem", The Journal of the Operational Research Society, 626–640, (1995).
• [52] J.M. Wilson, "Integer programming formulations of statistical classification problems", Omega-International Journal of Management Science, 24(6): 681–688, (1996).
• [53] T. Sueyoshi, "Mixed integer programming approach of extended DEA-Discriminant analysis", European Journal of Operational Research, 45–55: (2004).
• [54] T. Sueyoshi, "DEA-Discriminant analysis: Methodological comparison among eight discriminant analysis approaches", European Journal of Operational Research, 169: 247–272, (2006).
• [55] H. Bal, H.H. Örkcü, "A new mathematical programming approach to multi-group classification problems", Computers & Operations Research, 38: 105–111, (2011).
• [56] W. Gochet, A. Stam, V. Srinivasan, S. Chen, "Multigroup discriminant analysis using linear programming", Operations Research, 45(2): 213–225, (1997).
• [57] G. Xu, L.G. Papageorgiou, "A mixed integer optimisation model for data classification", Computers & Industrial Engineering, 56(4): 1205–1215, (2009).
• [58] F. Uney, M. Türkay, "A mixed-integer programming approach to multi-class data classification problem", European Journal of Operational Research., 173: 910–920, (2006).
• [59] L. Yang, S. Liu, S. Tsoka, L.G. Papageorgiou, "Sample re-weighting hyper box classifier for multi-class data classification", Computers & Industrial Engineering, 85: 44–56, (2015).
• [60] C. Cortes, V. Vapnik, "Support-Vector Networks", Machine Learning, 20: 273–297, (1995).
• [61] R.A. Fisher, "The use of multiple measurements in taxonomic problems", Annals of Eugenics. 7: 179–188, (1936).
• [62] J. Lakoumentas, J. Drakos, M. Karakantza, G. Sakellaropoulos, V. Megalooikonomou, G. Nikiforidis, "Optimizations of the naïve-Bayes classifier for the prognosis of B-Chronic lymphocytic leukemia incorporating flow cytometry data", Computers Methods and Programs in Biomedicine, 108(1): 158–167, (2012).
• [63] S.-K. Lee, P. Kang, S. Cho, "Probabilistic local reconstruction for k-NN regression and its application to virtual metrology in semiconductor manufacturing", Neurocomputing,. 131: 427–439, (2014). [64] R.G. Andrzejak, K. Lehnertz, F. Mormann, C. Rieke, P. David, C.E. Elger, "Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state", Physical Review:E, 64: (2001)
• [65]https://www.kaggle.com/datasets/harunshimanto/epileptic-seizure-recognition.
• [66] M.D. Latt, H.B. Menz, V.S. Fung, S.R. Lord, "Acceleration patterns of the head and pelvis during gait in older people with Parkinson’s disease: A comparison of fallers and nonfallers", Journal of Gerontology Series A, 64(6): 700–706, (2009).
• [67] M. Sekine, T. Tamura, M. Yoshida, Y. Suda, Y. Kimura, H. Miyoshi, Y. Kijima, Y. Higashi, T. Fujimoto, "A gait abnormality measure based on root mean square of trunk acceleration", Journal of Neuroengineering and Rehabilitation, 10(118): (2013).
• [68] E. de M. Mesquita, F.B. Rodrigues, A.P. Rodrigues, T.S. Lemes, A.O. Andrade, M.F. Vieira, "Discrimination capability of linear and nonlinear gait features in group classification", Medical Engineering & Physics, 93: 59–71, (2021).
• [69] J.H. Ko, K.M. Newell, "Aging and the complexity of center of pressure in static and dynamic postural tasks", Neuroscience Letters, 610: 104–109, (2016).
• [70] S. Ramdani, B. Seigle, J. Lagarde, F. Bouchara, P.L. Bernard, "On the use of sample entropy to analyze human postural sway data", Medical Engineering & Physics, 31(8): 1023–1031, (2009).