CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING

Budak, Ümit; AKBULUT, Yaman; ŞENGÜR, Abdulkadir

CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING

Abdulkadir ŞENGÜR, (Fırat Üniversitesi, Elektrik-Elektronik Mühendisliği Bölümü, Teknoloji Fakültesi, Elazığ, Türkiye)

Ümit BUDAK, (Bitlis Eren Üniversitesi, Mühendislik Fakültesi, Elektrik ve Elektronik Mühendisliği Bölümü, Bitlis, Türkiye)

Yaman AKBULUT (Fırat Üniversitesi, Bilişim Bölümü, Elazığ, Türkiye)

European Journal of Technique

0 0

Yıl: 2018 Cilt: 8 Sayı: 2 Sayfa Aralığı: 179 - 185 Metin Dili: İngilizce İndeks Tarihi: 14-11-2020

CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING

Öz:

This paper investigates the usage of transfer learning in amyotrophic lateralsclerosis (ALS) disease detection. ALS is a dangerous disease which affectsthe nerve cells in brain and spinal cord. Electromyogram (EMG) is animportant measure for analysing of the electrical level of the muscles. EMGbased early ALS disease detection system helps the physicians and patients.The proposed work uses EMG signals in discrimination of the ALS and healthypersons. The EMG signals are initially segmented with a overlapped windowand each segment is converted to the spectrogram images. The obtainedspectrogram images are resized and fed into the pre-trained convolutionalneural networks model. The pre-trained model is fine-tuned with the problemat hand. The R002 dataset which is obtained from www.emglab.net is usedduring the experimental works. Accuracy, sensitivity and specificity measuresare used to evaluate the obtained achievement. According to these measures,97.70% accuracy, 97.97% sensitivity, and 97.29% specificity values arerecorded. We further compare the obtained results with some of the existingresults that were obtained on the same dataset. The comparisons show thatproposed method is outperformed.

Anahtar Kelime:

Belge Türü: Makale Makale Türü: Araştırma Makalesi Erişim Türü: Erişime Açık

[1] Fuglsang‐ Frederiksen, A., The utility of interference pattern analysis, Muscle & Nerve: Official Journal of the American Association of Electrodiagnostic Medicine, 23(1), 2000, pp. 18-36.
[2] Fukuda, T. Y. et al., Root mean square value of the electromyographic signal in the isometric torque of the quadriceps, hamstrings and brachial biceps muscles in female subjects, Journal of Applied Research, 10(1), 2010, pp. 32-39.
[3] Fattah, S. A. et al., Evaluation of different time and frequency domain features of motor neuron and musculoskeletal diseases, Int. J. Comput. Appl., 43(23), 2012, pp. 34–40.
[4] Doulah, A. B. M. S. U., Fattah, S. A., Neuromuscular disease classification based on mel frequency cepstrum of motor unit action potential, In Electrical Engineering and Information & Communication Technology (ICEEICT), 2014 International Conference on IEEE, pp. 1-4.
[5] Mishra, V. K. et al., Analysis of ALS and normal EMG signals based on empirical mode decomposition, IET Science, Measurement & Technology, 10(8), 2016, pp. 963-971.
[6] Sengur, A. et al., DeepEMGNet: an application for efficient discrimination of ALS and normal EMG signals, T. Bvrezina, R. Jabłoński (Eds.), Mechatronics 2017 Recent Technol. Sc i. Adv., Springer International Publishing, Cham (2018), pp. 619-625, Doi: 10.1007/978-3-319-65960- 2_77.
[7] Fattah, S. A. et al., Identification of motor neuron disease using wavelet domain features extracted from EMG signal, In Circuits and Systems (ISCAS), International Symposium on IEEE, 2013, pp. 1308-1311.
[8] Pal, P. et al., Feature extraction for evaluation of Muscular Atrophy, In Computational Intelligence and Computing Research (ICCIC), International Conference on IEEE, December 2010, pp. 1-4.
[9] Merlo, A. et al., A fast and reliable technique for muscle activity detection from surface EMG signals, IEEE Transactions on Biomedical Engineering, 50(3), 20113, pp. 316-323.
[10] Sengur, A. et al., Classification of amyotrophic lateral sclerosis disease based on convolutional neural network and reinforcement sample learning algorithm, Health information science and systems, 5(1), 2017, 9. Doi: 10.1007/s13755-017-0029-6
[11] Orenstein, E. C., Beijbom, O., Transfer Learning and Deep Feature Extraction for Planktonic Image Data Sets, In Applications of Computer Vision (WACV), Winter Conference on IEEE, March 2017, pp. 1082-1088.
[12] Krizhevsky, A. et al., Imagenet classification with deep convolutional neural networks, In Advances in neural information processing systems, 2012, pp. 1097-1105.
[13] Nikolic, M., Detailed analysis of clinical electromyography signals: EMG decomposition, findings and firing pattern analysis in controls and patients with myopathy and amyotrophic lateral sclerosis, 2001, Doctoral dissertation.
[14] Simonyan, K., Zisserman, A., Very deep convolutional networks for large-scale image recognition, 2014, arXiv preprint arXiv:1409.1556.
[15] Deniz, E., et al., Transfer learning based histopathologic image classification for breast cancer detection, Health information science and systems, 2018, 6(1), 18.
[16] Detailed analysis of clinical electromyography signals EMG decomposi-tion, findings and firing pattern analysis in controls and patients with myopathy and amy-trophic lateral sclerosis. Ph.D. Thesis, Faculty of Health Science, University of Copenhagen.

APA	ŞENGÜR A, Budak Ü, AKBULUT Y (2018). CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING. , 179 - 185.
Chicago	ŞENGÜR Abdulkadir,Budak Ümit,AKBULUT Yaman CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING. (2018): 179 - 185.
MLA	ŞENGÜR Abdulkadir,Budak Ümit,AKBULUT Yaman CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING. , 2018, ss.179 - 185.
AMA	ŞENGÜR A,Budak Ü,AKBULUT Y CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING. . 2018; 179 - 185.
Vancouver	ŞENGÜR A,Budak Ü,AKBULUT Y CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING. . 2018; 179 - 185.
IEEE	ŞENGÜR A,Budak Ü,AKBULUT Y "CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING." , ss.179 - 185, 2018.
ISNAD	ŞENGÜR, Abdulkadir vd. "CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING". (2018), 179-185.

APA	ŞENGÜR A, Budak Ü, AKBULUT Y (2018). CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING. European Journal of Technique, 8(2), 179 - 185.
Chicago	ŞENGÜR Abdulkadir,Budak Ümit,AKBULUT Yaman CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING. European Journal of Technique 8, no.2 (2018): 179 - 185.
MLA	ŞENGÜR Abdulkadir,Budak Ümit,AKBULUT Yaman CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING. European Journal of Technique, vol.8, no.2, 2018, ss.179 - 185.
AMA	ŞENGÜR A,Budak Ü,AKBULUT Y CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING. European Journal of Technique. 2018; 8(2): 179 - 185.
Vancouver	ŞENGÜR A,Budak Ü,AKBULUT Y CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING. European Journal of Technique. 2018; 8(2): 179 - 185.
IEEE	ŞENGÜR A,Budak Ü,AKBULUT Y "CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING." European Journal of Technique, 8, ss.179 - 185, 2018.
ISNAD	ŞENGÜR, Abdulkadir vd. "CLASSIFICATION OF AMYOTROPHIC LATERAL SCLEROSIS AND HEALTHY ELECTROMYOGRAPHY SIGNALS BASED ON TRANSFER LEARNING". European Journal of Technique 8/2 (2018), 179-185.