Hybrid Brain-Computer Interface: a Novel Method on the Integration of EEG and sEMG Signal for Active Prosthetic Control

Authors

Reza Darmakusuma, Electrical Engineering, School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Bandung 40132, IndonesiaFollow
Ary Setijadi Prihatmanto, Electrical Engineering, School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Bandung 40132, Indonesia
Adi Indrayanto, Electrical Engineering, School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Bandung 40132, Indonesia
Tati Latifah Mengko, Electrical Engineering, School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Bandung 40132, Indonesia

Abstract

This paper describes a novel method for controlling active prosthetics by integrating surface electromyography (sEMG) and electroencephalograph signals to improve its intuitiveness. This paper also compares the new method (RTA-2) with other existing methods (AND and OR) for controlling active prosthetics. Based on analysis, RTA-2 features higher true positive rate (TPR) and balanced accuracy (BA) than AND method. On the other hand, the new method (RTA-2) yields lower false detection rate (FPR) than OR method. Analysis also shows that RTA-2 possesses equal TPR, FPR, and BA with the detection of movement intention using sEMG-based system. Although the RTA-2 method shows equal performance with the sEMG-based system, it presents an advantage for driving active prosthetics to move faster and to reduce its total time response by generating more movement commands.

Bahasa Abstract

Hybrid Brain-Computer Interface: Metode Baru dalam Integrasi Sinyal EEG dan sEMG untuk Pengendalian Aktif Prosthetics. Paper ini menjelaskan metode baru untuk mengendalikan prosthetics aktif dengan mengintegrasikan sinyal elektromiograf (sEMG) dan elektroensefalograf (EEG) dalam rangka meningkatkan sifat intuitif yang dimilikinya. Selain itu, dalam paper ini juga membandingkan metode baru (RTA-2) dengan metode lain yang telah ada (AND dan OR) untuk mengendilikan prosthetics aktif. Berdasarkan analisis, metode RTA-2 memiliki nilai True Positive Rate (TPR) dan Balanced Accuracy (BA) lebih tinggi dibandingkan metode AND. Selain hal terebut, metode RTA-2 memilki kesalahan deteksi (FPR) yang lebih rendah dibandingkan metode OR. Berdasarkan analasis, nilai TPR, FPR dan BA yang dimiliki metode RTA-2 ini sama dengan akurasi deteksi intensi gerakan berbasis sinyal sEMG. Namun demikian, meskipun TPR, FPR dan BA dari metode RTA-2 sama dengan metode yang hanya berbasis sinyal sEMG, metode RTA-2 memiliki keunggulan dalam mengendalikan prosthetic aktif sehingga dapat bergerak dengan kecepatan lebih cepat dari sebelumnya dan mengurangi total waktu responnya dengan cara menghasilkan perintah keluaran kecepatan gerakan yang lebih banyak.

References

1. O.Z. Khokhar, Z.G. Xiao, C. Menon, Biomed. Eng.
2. Online. 9 (2010) 41.
3. A. Fougner, Ph.D Thesis, Faculty of Information Technology, Mathematics and Electrical Engineering, Department of Engineering Cybernetics, Norwegian University of Science and Technology, 2013.
4. G. Li, Electromyography Pattern-Recognition-Based Control of Powered Multifunctional Upper-Limb Prostheses, Advances in Applied Electromyography, Prof. Joseph Mizrahi (Ed.), InTech, 2011, p.99.
5. S. Herle, S. Man, Processing Surface Electro-myographical Signals for Myoelectric Control, Rehabilitation Engineering. Tan Yen Kheng (Ed.), InTech, 2009, p.223.
6. E.A. Kirchner, M. Tabie. A. Seeland, PLoS ONE, 9 (2014) 1:e85060.
7. E. Lew, R. Chavarriaga, H. Zhang, M. Seeck, J.R. Millan, Self-paced Movement Intention Detection from Human Brain Signals: Invasive and Non-invasive EEG, Proceedings of 34th Annual International Conference of EMBS, IEEE, San Diego, California, USA., 2012, p.3280.
8. H. Shibasaki, M. Harlett, J. Clin. Neurophysiol. 117 (2006) 2341.
9. D. Planelles, E. Hortal, A. Costa, A. Ubeda, E. Ianez, J.M. Azorín, J. Sensors. 14 (2014) 18172.
10. R. Leeb, H. Sagha, R. Chavarriaga, J.R. Millan, J. Neural Eng. 8/2 (2011) 025011.
11. B. Meikle, M. Devlin, T. Pauley, J. Phys. Med. Rehabil. 87/3 (2006) 390.
12. R. Darmakusuma, A.S. Prihatmanto, A. Indrayanto, T.L. Mengko, Far East J. Electron. Commun. 16/2 (2016) 381.
13. A. Fougner, Ø. Stavdahl, P.J. Kyberd, Y.G. Losier, P.A. Parker, IEEE Trans. Neural Syst. Rehabil. Eng. 20/5 (2012) 663.
14. D.S. Childress. J. Clin, Orthot. Int. 9/1 (1985) 2.
15. O. Bai, P. Lin, S. Vorbach, Floeter, K. Mary, N. Hattori, M. Hallett, J. Neural Eng. 5 (2008) 24.
16. J. Jones, Statistics: Lecture Notes, https://people. richland.edu/james/lecture/m170/ch05-rul.html, 1996.
17. J. Feher, Quantitative Human Physiology: An Introduction, Academic Press Series in Biomedical Engineering, Elsevier, New York, 2012, p.341.
18. R. Darmakusuma, A.S. Prihatmanto, A. Indrayanto, T.L. Mengko, L.A. Andarini, A.F. Idrus. Makara J. Technol. 20/1 (2016) 38.
19. G. Schalk, J. Mellinger, User Tutorial: Introduction to The Mu Rhythm, http://www.bci2000.org, 2008.
20. A. Panagiotis, Neuro-Robotics: From Brain Machine Interfaces to Rehabilitation Robotics, Vol. 2, Trends in Augmentation of Human Performance, Springer, Berlin, 2014.

Recommended Citation

Darmakusuma, Reza; Prihatmanto, Ary Setijadi; Indrayanto, Adi; and Mengko, Tati Latifah (2018) "Hybrid Brain-Computer Interface: a Novel Method on the Integration of EEG and sEMG Signal for Active Prosthetic Control," Makara Journal of Technology: Vol. 22: Iss. 1, Article 5.
DOI: 10.7454/mst.v22i1.3103
Available at: https://scholarhub.ui.ac.id/mjt/vol22/iss1/5