Bağımsız Bileşen Analizi ve Makine Öğrenmesi Kullanılarak Omurilik Yaralanması Olan Kişilerden Alınan EEG Sinyallerinden El Hareketlerinin Sınıflandırılması

Year 2024, Volume: 14 Issue: 3, 1225 - 1244, 15.09.2024

Ebru Sayılgan

https://doi.org/10.31466/kfbd.1447072

Abstract

Bu çalışmanın temel amacı, Omurilik Yaralanması (OY) olan kişilerin, kol ve el hareketlerinin, kodu çözülebilir nöral bağıntılarını koruduğunu göstermektir. OY’li on kişiden pronasyon, süpinasyon, palmar kavrama, lateral kavrama ve el açma hareketleri düşündürülerek alınan ElektroEnsefaloGrafi (EEG) sinyallerinin ayırt edici hareket bilgisi araştırılmıştır. Bunu yaparken kullanılan yöntemlerde Bağımsız Bileşen Analizi (BBA/ICA) yöntemi hem artefakt gidermede hem de yeni bir yaklaşım olarak öznitelik vektörlerini çıkarmada kullanılmıştır. Önerilen yöntemde öznitelik vektörleri bağımsız bileşenlerde ortak bilgi matrisi çıkarılarak oluşturulmuştur. Çıkarılan ve seçimi yapılan öznitelik vektörleri dört farklı makine öğrenmesi modeli (Destek Vektör Makinesi (DVM), k-En Yakın Komşuluk (k-EYK), AdaBoost ve Karar Ağaçları (KA)) ile test edilmiştir. Model değerlendirme aşamasında aşırı öğrenmenin önüne geçmek için 5-katlamalı çapraz doğrulama ve hata matrisi yöntemleri kullanılmıştır. Sonuç olarak, incelenen beş sınıfa göre elde edilen başarım oldukça yüksek çıkmıştır. Deneklerin ortalaması alındığında elde edilen model doğruluk değerleri sırasıyla DVM’de 0.9024±0.0781, k-EYK’da 0.8582±0.0985, AdaBoost’ta 0.7924±0.0937 ve KA’da 0.8089±0.0645 olarak hesaplanmıştır. Bu sonuçlara dayanarak OY olan bireylerin kol ve el hareketlerinin ayırt edicilik performansının önerilen yöntem ile oldukça yüksek sonuçlar verdiği görülmektedir. BBA yöntemine dayalı bir öznitelik çıkarma ve DVM modeli ile sınıflandırma metodolojisinin OY’li hastaların rehabilitasyon tedavisinde EEG temelli beyin bilgisayar arayüzü uygulamalarına önemli bir katkısı olacağı düşünülmektedir.

Keywords

Omurilik yaralanması, EEG, Bağımsız bileşen analizi, Sınıflandırma, Makine öğrenmesi

Classification of Hand Movements from EEG Signals from Persons with Spinal Cord Injury Using Independent Component Analysis and Machine Learning

Abstract

The main aim of this study is to demonstrate that people with Spinal Cord Injury (SCI) preserve decipherable neural correlates of arm and hand movements. The distinctive movement information of ElectroEncephaloGraphy (EEG) signals obtained by imagining pronation, supination, palmar grasp, lateral grasp and hand open movements from eight subjects with SCI was investigated. In doing so, Independent Component Analysis (ICA) was used in both artifact removal and feature vector extraction as a new approach. In the proposed method, feature vectors were created by extracting the common information matrix in independent components. Extracted and selected feature vectors were tested with four different machine learning models (Support Vector Machine (SVM), k-Nearest Neighbor (kNN), AdaBoost, and Decision Trees (DT)). In the model evaluation phase, 5-fold cross validation and confusion matrix methods were used to prevent over-learning. As a result, according to the five classes the obtained performance is quite high. Depending on all subject's average values, the model accuracies were calculated as 0.9024±0.0781 in SVM, 0.8582±0.0985 in kNN, 0.7924±0.0937 in AdaBoost, and 0.8089±0.0645 in DT, respectively. Based on these results, it is seen that the discrimination performance of arm and hand movements of individuals with SCI is high with the proposed method. It is thought that a feature extraction and classification methodology based on the ICA method and SVM model will make an important contribution to EEG-based brain computer interface applications in the rehabilitation treatment of patients with SCI.

Keywords

Spinal cord injury, EEG, Independent component analysis, classification, Machine learning

References

• Agarwal, S., Zubair, M. (2021). Classification of Alcoholic and Non-Alcoholic EEG Signals based on Sliding-SSA and Independent Component Analysis. IEEE Sensors Journal, 21(23), 26198-26206. doi: 10.1109/JSEN.2021.3120885
• Akman Aydın, E. (2023). Detection of Movement Related Cortical Potentials from Single Trial EEG Signals. Gazi University Journal of Science Part C: Design and Technology, 11(1): 25-38. doi: 10.29109/gujsc.1083912
• Athanasiou, A., Klados, M. A., Pandria, N., Foroglou, N., Kavazidi, K. R., Polyzoidis, K., Bamidis, P. D. (2012). A Systematic Review of Investigations into Functional Brain Connectivity Following Spinal Cord Injury. Frontiers in Human Neuroscience, 11, 517. doi: 10.3389/fnhum.2017.00517
• Bascil, M. S., Tesneli, A. Y., Temurtas, F. (2016). Spectral Feature Extraction of EEG Signals and Pattern Recognition During Mental Tasks of 2-D Cursor Movements for BCI Using SVM and ANN. Australasian Physical and Engineering Sciences in Medicine, 39, 665–676. doi: 10.1007/s13246-016-0462-x
• Bell, A. J., Sejnowski, T. J. (1995). An Information-Maximization Approach to Blind Separation and Blind Deconvolution. Neural Computation, 7(6), 1129–1159.
• Cancino, S., López, J. M., Delgado Saa, J. F., Schettini, N. (2023). ConvNets for Electroencephalographic Decoding of Attempted Arm and Hand Movements of People with Spinal Cord Injury. Advanced Intelligent System, 5(12): 2023. doi: 10.1002/aisy.202300094
• Cao, C., Slobounov, S. (2010). Alteration of Cortical Functional Connectivity As A Result of Traumatic Brain Injury Revealed by Graph Theory, ICA, and sLORETA Analyses of EEG Signals. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 18(1), 11-19. doi: 10.1109/TNSRE.2009.2027704
• Chaudhary, S., Taran, S., Bajaj, V., Siuly, S. (2020). A Flexible Analytic Wavelet Transform Based Approach For Motor-Imagery Tasks Classification In BCI Applications. Computer Methods and Programs in Biomedicine, 187(2020), 105325. doi: 10.1016/j.cmpb.2020.105325
• Chaumon, M., Bishop, D. V., Busch, N. A. (2015). A Practical Guide to The Selection of Independent Components of The Electroencephalogram for Artifact Correction. Journal of Neuroscience Methods, 250, 47–63.
• Demir, A., Bekiryazıcı, Ş., Coşkun, O., Eken, R., Yılmaz, G. (2022). Detection and Analysis of Driver Fatigue Stages with EEG Signals. Pamukkale Universitesi Mühendislik Bilimleri Dergisi, 28(5), 643-651.
• Delorme, A., Makeig, S. (2004). EEGLAB: An Open-Source Toolbox for Analysis of Single-Trial EEG Dynamics Including Independent Component Analysis. Journal of Neuroscience Methods, 134(1), 9–21.
• Dev, A., ve ark. (2022). Exploration of EEG-Based Depression Biomarkers Identification Techniques and Their Applications: A Systematic Review. IEEE Access, 10, 16756-16781. doi: 10.1109/ACCESS.2022.3146711
• Hernandez-Rojas, L. G., ve ark. (2022). Brain-Computer Interface Controlled Functional Electrical Stimulation: Evaluation with Healthy Subjects and Spinal Cord Injury Patients. IEEE Access, 10, 46834-46852. doi: 10.1109/ACCESS.2022.3170906
• Jung, Y. (2018). Multiple Predicting k-Fold Cross-Validation for Model Selection. Journal of Nonparametric Statistics, 30(1), 197-215. doi: 10.1080/10485252.2017.1404598
• Kang, J. S., Kavuri, S., Lee, M. (2022). ICA-Evolution Based Data Augmentation with Ensemble Deep Neural Networks Using Time and Frequency Kernels for Emotion Recognition from EEG-Data. IEEE Transactions on Affective Computing, 13(2), 616-627. doi: 10.1109/TAFFC.2019.2942587
• Kang, Y., Ding, H., Zhou, H., Wei, Z., Liu, L., Pan, D. Y., Feng, S. Q. (2018). Epidemiology of Worldwide Spinal Cord Injury: A Literature Review. Neurorestoratology, 6, 1–9. doi: 10.2147/jn.s143236
• Kavuri, S. S., Veluvolu, K. C., Chai, Q. H. (2018). Evolutionary Based ICA with Reference for EEG μ Rhythm Extraction. IEEE Access, 6, 19702-19713. doi: 10.1109/ACCESS.2018.2821838
• Khoshnevis, S. A., Sankar, R. (2020). Applications of Higher Order Statistics in Electroencephalography Signal Processing: A Comprehensive Survey. IEEE Reviews in Biomedical Engineering, 13, 169-183. doi: 10.1109/RBME.2019.2951328
• Kotsiantis, S. B. (2013). Decision Trees: A Recent Overview. Artif Intell Rev, 39, 261–283. doi: 10.1007/s10462-011-9272-4
• Leite, V. F., deSouza, D. R., Imamura, M., Batistella, L. R. (2019). Post-discharge Mortality in Patients with Traumatic Spinal Cord Injury in A Brazilian Hospital: A Retrospective Cohort. Spinal Cord, 57, 134–140. doi: 10.1038/s41393-018-0183-y
• Liu, J., Yang, X., Jiang, L., Wang, C., Yang, M. (2012). Neural Plasticity After Spinal Cord Injury. Neural Regeneration Research, 7, 386–391.
• Makouei, S.T.Z., Uyulan, Ç. (2023). Classification of the Attempted Arm and Hand Movements of Patients with Spinal Cord Injury Using Deep Learning Approach. doi: 10.1101/2023.07.06.23292320
• Mirzabagherian, H., Menhaj, M. B., Suratgar, A. A., Talebi, N., Abbasi Sardari, M. R., Sajedin, A. (2023). Temporal-Spatial Convolutional Residual Network for Decoding Attempted Movement Related EEG Signals of Subjects with Spinal Cord Injury. Computers in Biology and Medicine, 164(2023), 107159. doi: 10.1016/j.compbiomed.2023.107159
• Mohseni, M., Shalchyan, V., Jochumsen, M., Niazi, I. K. (2020). Upper Limb Complex Movements Decoding from Pre-Movement EEG Signals Using Wavelet Common Spatial Patterns. Computer Methods and Programs in Biomedicine, 183 (2020), 105076. doi: 10.1016/j.cmpb.2019.105076
• Nam, K. Y., Kim, H. J., Kwon, B. S., Park, J. W., Lee, H. J., Yoo, A. (2017). Robot-Assisted Gait Training (Lokomat) Improves Walking Function and Activity in People with Spinal Cord Injury: A Systematic Review. Journal of NeuroEngineering and Rehabilitation, 14(1), 24.
• Ohsaki, M., Wang, P., Matsuda, K., Katagiri, S., Watanabe, H., Ralescu, A. (2017). Confusion-Matrix-based Kernel Logistic Regression for Imbalanced Data Classification. IEEE Transactions on Knowledge and Data Engineering, 29(9), 1806-1819. doi: 10.1109/TKDE.2017.2682249
• Ofner, P., Schwarz, A., Pereira, J., ve ark. (2019). Attempted Arm and Hand Movements Can Be Decoded from Low-Frequency EEG from Persons with Spinal Cord Injury. Scientific Reports, 9, 7134. doi: 10.1038/s41598-019-43594-9
• Ofner, P., Schwarz, A., Pereira, J., Muller-Putz, G. R. (2017). Upper Limb Movements Can Be Decoded from The Time-Domain of Low-Frequency EEG. PLoS One, 12 (2017), e0182578. doi: 10.1371/journal.pone.0182578
• Papana, A., Kugiumtzis, D. (2009). Evaluation of Mutual Information Estimators for Time Series. International Journal of Bifurcation and Chaos, 19(12), 4197–4215.
• Pfurtscheller, G., Linortner, P., Winkler, R., Korisek, G., Muller-Putz G. (2009). Discrimination of Motor Imagery-Induced EEG Patterns in Patients with Complete Spinal Cord Injury. Computational Intelligence and Neuroscience, 104180. doi: 10.1155/2009/104180
• Sai, C. Y., Mokhtar, N., Arof, H., Cumming, P., Iwahashi, M. (2018). Automated Classification and Removal of EEG Artifacts with SVM and Wavelet-ICA. IEEE Journal of Biomedical and Health Informatics, 22(3), 664-670. doi: 10.1109/JBHI.2017.2723420
• Sayilgan, E., Yuce, Y. K., Isler, Y. (2021). Evaluation of Mother Wavelets on Steady-State Visually-Evoked Potentials for Triple-Command Brain-Computer Interfaces. Turkish Journal of Electrical Engineering and Computer Sciences, 29(5), 2263-2279.
• Sayılgan, E., Yüce, Y., İşler, Y. (2021). Uyartım Frekansının Kestiriminde İstatistiksel Anlamlılığa Dayalı Olarak Seçilen Durağan Durum Görsel Uyarılmış Potansiyellere Ait Dalgacık Özniteliklerinin Değerlendirilmesi. Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, 36(2), 593-606. doi: 10.17341/gazimmfd.664583
• Sayilgan, E., Yuce, Y. K., Isler, Y. (2022). Investigating The Effect of Flickering Frequency Pair and Mother Wavelet Selection in Steady-State Visually-Evoked Potentials on Two-Command Brain-Computer Interfaces. Innovation and Research in BioMedical engineering, 43(6), 594-603. doi: 10.1016/j.irbm.2022.04.006
• Schapire, R. E. (2013). Explaining AdaBoost. Editors: Schölkopf B, Luo Z, Vovk V. Empirical Inference. Berlin, Heidelberg, Springer. doi: 10.1007/978-3-642-41136-6_5
• Schlogl, A., Lee, F., Bischof, H., Pfurtscheller, G. (2005). Characterization of Four-Class Motor Imagery EEG Data for The BCI-Competition 2005. Journal of Neural Engineering, 2 (2005): L14–22. doi: 10.1088/1741-2560/2/4/L02
• Simis, M., Uygur-Kucukseymen, E., Pacheco-Barrios, K., Battistella, L. R., Fregni, F. (2020). Beta-Band Oscillations As A Biomarker of Gait Recovery in Spinal Cord Injury Patients: A Quantitative Electroencephalography Analysis. Clinical Neurophysiology, 131, 1806–1814.
• Sreeja, S.R., Samanta, D. (2019). Classification of Multiclass Motor Imagery EEG Signal Using Sparsity Approach. Neurocomputing, 368 (2019): 133–145. doi: 10.1016/j.neucom.2019.08.037
• Suyal, M., Goyal, P. (2022). A Review on Analysis of k-Nearest Neighbor Classification Machine Learning Algorithms Based on Supervised Learning. International Journal of Engineering Trends and Technology, 70(7), 43-48. doi: 10.14445/22315381/IJETT-V70I7P205
• Wang, T., Deng, J., He, B. (2004). Classifying EEG-Based Motor Imagery Tasks By Means Of Time-Frequency Synthesized Spatial Patterns. Clinical Neurophysiology, 115 : 2744–2753. doi: 10.1016/j.clinph.2004.06.022
• Wang, Y., Jung, T. P. (2012). Improving Brain–Computer Interfaces Using Independent Component Analysis. In: Allison, B., Dunne, S., Leeb, R., Del R. Millán, J., Nijholt, A. (eds) Towards Practical Brain-Computer Interfaces. Biological and Medical Physics, Biomedical Engineering. Berlin, Heidelberg, Springer. doi: 10.1007/978-3-642-29746-5_4
• Wong, T. T., Yeh, P. Y. (2020). Reliable Accuracy Estimates from k-Fold Cross Validation. IEEE Transactions on Knowledge and Data Engineering, 32(8), 1586-1594. doi: 10.1109/TKDE.2019.2912815
• Zhou, X., Zou, R., Huang, X. (2021). Single Upper Limb Functional Movements Decoding from Motor Imagery EEG Signals Using Wavelet Neural Network. Biomedical Signal Processing and Control, 70(2021), 102965. doi: 10.1016/j.bspc.2021.102965