ESTIMATION OF PAIN THRESHOLD FROM EEG SIGNALS OF SUBJECTS IN PHYSICAL THERAPY USING LONG-SHORT-TERM MEMORY DEEP LEARNING MODEL

Year 2021, Volume: 26 Issue: 2, 447 - 460, 31.08.2021

Kutay Güneç Ömer Kasım Mustafa Tosun Emine Büyükköroğlu

https://doi.org/10.17482/uumfd.883100

Cited By: 3

Abstract

Pain is a natural stimulation to protect the whole body. An overreaction to pain can damage the tissues. Therefore, it is important to know the angle at which pain is felt when routinely measuring joint range of motion during the first examination. Detection of pain with the change in characteristics of electroencephalogram signals at the moments when pain occurs is the novelty of this study. The characteristics of the signal with power band changes were obtained by frequency analysis of the electroencephalogram signals. Pain was detected by classifying these characteristics with the Long Short Term Memory deep learning model. Validation of the model was performed with records obtained from 43 volunteer subjects with a 14-channel wireless Emotive brand electroencephalogram device. 96.1% success in binary classification as with pain or without pain and 89.6% success in multi-class classification as with high pain, low pain and without pain was achieved. This success is a quality that can support specialists in diagnosis and treatment by determining the threshold where pain occurs during the first physical therapy examination from the electroencephalogram signals.

Keywords

Pain Threshold, EEG Signal Processing, Power Spectrum, LSTM Deep Learning Model

Fizik Tedavide Hastaların EEG Sinyallerinden Ağrı Eşiğinin Uzun Kısa Süreli Hafıza Derin Öğrenme Modeliyle Kestirimi

Abstract

Ağrı, tüm vücudu korumak için doğal bir uyarıdır. Bu uyarıya karşı gösterilecek aşırı reaksiyon, dokuda hasarlara neden olmaktadır. İlk muayenede rutin olarak eklem hareket açıklığı (EHA) ölçümünde ağrının hissedildiği açının bilinmesi önemlidir. Ağrının oluştuğu anlardaki EEG sinyallerindeki güç değişimi ile ağrının tespiti bu çalışmanın yeniliğidir. EEG sinyallerinin frekans analizi ile güç bandı değişimleri ile sinyale ait özellikler elde edilmiştir. Bu özellikler LSTM derin öğrenme modeli ile sınıflandırılarak ağrı tespit edilmiştir. Modelin doğrulanması bu çalışma kapsamında 43 gönüllü hastadan, 14 kanallı kablosuz Emotive marka EEG cihazıyla alınan kayıtlar ile yapılmıştır. İkili sınıflandırmada %96,1 çoklu sınıflandırmada ise %89,6’lik başarı elde edilmiştir. Bu başarı, ilk fizik tedavi muayenesi sırasında ağrının oluştuğu eşiğin EEG sinyallerinden belirlemesiyle uzmanları tanı ve tedavide destekleyebilecek bir niteliktir.

Keywords

Ağrı Eşiği, EEG Sinyal İşleme, Güç Spektrumu, LSTM Derin Öğrenme Modeli

References

• Camfferman, D., Moseley, G. L., Gertz, K., Pettet, M. W., Jensen, M. P. (2017) Waking EEG cortical markers of chronic pain and sleepiness, Pain Medicine, 18(10), 1921-1931. doi:10.1093/pm/pnw294
• Cao Z, Lai K.L., Lin C.T., Chuang C.H., Chou C.C., Wang S.J. (2018) Exploring the complexity of resting state EEG before migraine attacks, Cephalalgia, 38 (7): 1296-1306. doi:10.1177/0333102417733953
• Cao, T., Liu, D., Wang, Q., Tao, L., Sun, J. (2020) Frequency-Domain EEG Analysis for Sudden Pain Perception, IEEE International Conference on Artificial Intelligence and Information Systems, Dalian, China , 434-440. doi:10.1109/ICAIIS49377.2020.9194928
• Chen, Z., Zhang, Q., Tong, A. P. S., Manders, T. R., Wang, J. (2017) Deciphering neuronal population codes for acute thermal pain, Journal of neural engineering, 14(3), 036023. doi:10.1088/1741-2552/aa644d
• Das, P. and Babadi, B. (2020) Multitaper spectral analysis of neuronal spiking activity driven by latent stationary processes, Signal Processing, 170, 107429. doi:10.1016/j.sigpro.2019.107429
• Emotiv, E. P. O. C. (2014). Brain-Computer Interface and scientific contextual EEG. EMOTIV EPOC and testbench specifications,” EMOTIV Systems. Available at: http://emotiv.com/files/Emotiv-EPOC-Product-Sheet-2014.pdf. [Access:18-Tem 2020].
• Ertam, F. (2019) An effective gender recognition approach using voice data via deeper LSTM networks. Applied Acoustics, 156, 351-358. doi:10.1016/j.apacoust.2019.07.033
• Gross, J., Schnitzler, A., Timmermann, L., Ploner, M. (2007) Gamma oscillations in human primary somatosensory cortex reflect pain perception, PLoS Biol, 5(5), e133. doi:10.1371/journal.pbio.0050133
• Homan, R. W., Herman, J., Purdy, P. (1987) Cerebral location of international 10–20 system electrode placement, Electroencephalography and clinical neurophysiology, 66(4), 376-382. doi:10.1016/0013-4694(87)90206-9
• Hu, X., Yuan, S., Xu, F., Leng, Y., Yuan, K., & Yuan, Q. (2020) Scalp EEG classification using deep Bi-LSTM network for seizure detection, Computers in Biology and Medicine, 124, 103919. doi:10.1016/j.compbiomed.2020.103919
• Kara, A. Uzun-Kısa Süreli Bellek Ağı Kullanarak Global Güneş Işınımı Zaman Serileri Tahmini, Gazi Üniversitesi Fen Bilimleri Dergisi Part C: Tasarım ve Teknoloji, 7(4), 882-892. doi:10.29109/gujsc.571831
• Kim, H., Neubert, J. K., Rowan, J. S., Brahim, J. S., Iadarola, M. J., Dionne, R. A. (2004) Comparison of experimental and acute clinical pain responses in humans as pain phenotypes, The Journal of Pain, 5(7), 377-384. doi:10.1016/j.jpain.2004.06.003
• Kisler, L. B., Kim, J. A., Hemington, K. S., Rogachov, A., Cheng, J. C., Bosma, R. L., Davis, K. D. (2020) Abnormal alpha band power in the dynamic pain connectome is a marker of chronic pain with a neuropathic component, NeuroImage: Clinical, 102241. doi:10.1016/j.nicl.2020.102241
• Li, J., Dimitrakopoulos, G. N., Thangavel, P., Chen, G., Sun, Y., Guo, Z., Bezerianos, A. (2019). What Are Spectral and Spatial Distributions of EEG-EMG Correlations in Overground Walking? An Exploratory Study, IEEE Access, 7, 143935-143946. doi:10.1016/j.nicl.2020.102241
• Li, M. W., Geng, J., Hong, W. C., Zhang, L. D. (2019) Periodogram estimation based on LSSVR-CCPSO compensation for forecasting ship motion. Nonlinear Dynamics, 97(4), 2579-2594. doi:10.1007/s11071-019-05149-5
• Nath, D., Singh, M., Sethia, D., Kalra, D., & Indu, S. (2020) An Efficient Approach to EEG-Based Emotion Recognition using LSTM Network, 16th IEEE International Colloquium on Signal Processing & Its Applications, Langkawi Island, Malaysia, 88-92. doi:10.1109/CSPA48992.2020.9068691
• Özmen, N. G., Durmuş, E., Sadreddini, Z. (2017) Müzik Sınıflandırması Beyin Bilgisayar Arayüzü Uygulamaları İçin Bir Alternatif Olabilir Mi?, Uludağ University Journal of the Faculty of Engineering, 22(2), 11-22. doi:10.17482/uumfd.335419
• Panavaranan, P. and Wongsawat, Y. (2013) EEG-based pain estimation via fuzzy logic and polynomial kernel support vector machine, 6th Biomedical Engineering International Conference, Amphur Muang, Krabi, Thailand, 1-4. doi:10.1109/BMEiCon.2013.6687668
• Prichep, L. S., Shah, J., Merkin, H., Hiesiger, E. M. (2018) Exploration of the pathophysiology of chronic pain using quantitative EEG source localization, Clinical EEG and Neuroscience, 49(2), 103-113. doi:10.1177/1550059417736444
• Saeedi, A., Saeedi, M., Maghsoudi, A., & Shalbaf, A. (2020) Major depressive disorder diagnosis based on effective connectivity in EEG signals: A convolutional neural network and long short-term memory approach, Cognitive Neurodynamics, 1-14. doi:10.1007/s11571-020-09619-0
• Schulz, E., May E.S., Postorino, M., Tiemann, L., Nickel, M.M., Witkovsky, V., Schmidt, P., Gross, J., Ploner, M. (2015) Prefrontal gamma oscillations encode tonic pain in humans, Cereb Cortex, 25 (11) 4407–4414. doi:10.1093/cercor/bhv043
• Vanneste, S., Song, J. J., De Ridder, D. (2018) Thalamocortical dysrhythmia detected by machine learning, Nature communications, 9(1), 1-13. doi:10.1038/s41467-018-02820-0
• Veerbeek, J. M., Wegen, E., Peppen, R., Wees, P. J., Hendriks, E., Rietberg, M., Kwakkel, G. (2014) What is the evidence for physical therapy poststroke? A systematic review and meta-analysis, PloS one, 9(2), e87987. doi:10.1371/journal.pone.0087987
• Yu, M., Sun, Y., Zhu, B., Zhu, L., Lin, Y., Tang, X, Dong, M. (2020) Diverse frequency band-based convolutional neural networks for tonic cold pain assessment using EEG, Neurocomputing, 378, 270-282. doi:10.1016/j.neucom.2019.10.023