Home / Technology Policy & Ethics / September 2019 / Artificial Intelligence Powered EEG-EMG Electrodes for Assisting Paralyzed

Artificial Intelligence Powered EEG-EMG Electrodes for Assisting the Paralyzed

Sunil Jacob, Center for Robotics, Varun G Menon, Department of Computer Science and Engineering and Saira Joseph, Department of Electronics and Communication Engineering, SCMS School of Engineering and Technology, India

September 2019


Paralysis is the loss of muscle function in any part of a person’s body. This occurs when the passage of messages between the brain and muscles is hindered. Over the past few years, it is seen that paralysis is more widespread than ever before; a recent survey shows that around the world, 1 out of 111 people are affected by paralysis1,2. This raises the need for developing an interface to aid people who suffer from this unfortunate condition3,4. Paralysis limits a person from completing even basic chores in life without any assistance. At times, the paralyzed parts become so stiff, it is difficult for the caregivers to provide assistance. Physiotherapy is also not the best option because of its various limitations such as the need for multiple weekly appointments, long sessions and high cost. Most of our previous work had focused on designing of efficient exoskeletons for rehabilitation. But exoskeletons5,6  introduced additional burdens to patients with their heaviness and complexity to the caregivers. In this article we discuss an efficient and better way by which friends and caregivers can assist the paralyzed to improve their lifestyle. The primary objective of this article is to discuss the working of Artificial Intelligence based Electroencephalograph (EEG) – Electromyogram (EMG) electrodes for Paralyzed (AI-EEP). This technique has been recently investigated by our team at the Center for Robotics laboratory at SCMS School of Engineering and Technology, Cochin, India. This device will help paralyzed people to move independently by using the recorded movements of a normal person as a reference.

Working of the AI-EEP System

The working of the system is divided into two phases. In the first phase of the operation, the EEG signal of a normal person is recorded during voluntary movement using an EEG headset. This signal is fed as the input to the AI-EEP device and is used to set a threshold. The device then converts the EEG signals into corresponding two-dimensional movements, and the movements are classified based on the signal attributes. The accuracy of this classification is measured as the number of correct classifications divided by the number of physical movements.

In the second phase, the nerves responsible for the movement of specific muscles of the paralyzed person are identified. On these identified locations, EMG electrodes are placed and a Transcutaneous Electrical Nerve Stimulator (TENS) generates the desired nerve stimulation based on the recorded EEG signal. Performance accuracy is improved by a bio-feedback system.

System Architecture

The system architecture diagram is included as figure 1 in the article for better understanding. The device consists primarily of three modules:

  1. Recording module: The EEG signals taken from the EEG headset are first acquired by the AI-EEP device. A high gain instrumentation amplifier increases the strength of the impulses, and a band pass filter is used to select the desired range of frequencies before passing it to the conditioning module for preconditioning.
  2. Data preconditioning module: In this module the signal features are extracted after noise reduction, filtering and windowing and desired attributes of the signal are selected.
  3. Categorizing module: This module categorizes various EEG signals into pre-determined muscle movement segments. The EEG signals first trigger the TENS device. The TENS device in turn, generates the nerve stimulation and with the help of the EMG section, we get the desired muscle movement. Neuron activity of the brain differs from person to person; the EEG signals of one person may not work for another. Also, the EEG signals of a person for a particular thought will differ from time to time. Hence adaptability is a major issue in our project. The human muscle should adapt to the interface. For this, the decoded signal for muscle selection is obtained by a bio-feedback from the TENS device.

The mapping of EEG to EMG signal is a time-consuming process and continuous practice is needed to achieve the required accuracy. In order to overcome this problem, this technique uses a Self-Adaptive Algorithm (SAA). The SAA is a feedback and PID powered controller. With the Intelligence enabled system, it analyzes the received EEG signal and depending on the threshold, decides the contraction or expansion required for the muscle.  The set threshold locks the desired movement.

Figure 1: Architecture of the proposed system depicting the flow of signals from the recording module to the pre conditioning module and then to the categorizing module


The AI-EEP is a pioneering technology that can generate movement in the muscle of paralyzed people depending on EEG signals. It is a very efficient alternative to the existing exoskeletons. With exoskeletons, users have to carry the excess weight of the skeleton, and it has to be customized for different parts of the body. In AI-EEP there is no add-on weight of the system and the AI-EEP can be placed on any part of the body without any modification. This technique would be highly beneficial in assisting paralyzed people and improving their life style.

Future Research Directions

This work can be extended by enabling several EEG electrodes over a region of the brain. These electrodes will be controlled by a Node-MCU board and will also be IoT enabled. This will help to excite specific parts of the brain for the required movement. If the normal person is physically moving, then the paralyzed person enabled with AI-EEP will get activation signals only in the motor region of the brain. Similarly, we can activate desired regions of the brain and not excite all the electrodes in the headset. More research could be carried out to improve the classification accuracy of the EEG signals. Deep learning technology could also be used further to improve the efficiency of the system.

Latest Standards

The Neurophysiology EEG standard published in April 2016 Version 21 published by the College of Physicians and Surgeons of Alberta is the reference for the study of EEG measurement and its analysis7.


  1. B. Armour, E. Courtney-Long, M. Fox, H. Fredine and A. Cahill, “Prevalence and Causes of Paralysis—United States, 2013”, American Journal of Public Health, vol. 106, no. 10, pp. 1855-1857, 2016. DOI: 10.2105/ajph.2016.303270.
  2. Christopher & Dana Reeve Foundation. Paralysis statistics. https://www.christopherreeve.org/living-with-paralysis/stats-about-paralysis [Accessed on March 5, 2019].
  3. Q. Wu, X. Wang, B. Chen and H. Wu, “Development of a Minimal-Intervention-Based Admittance Control Strategy for Upper Extremity Rehabilitation Exoskeleton,” in IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 48, no. 6, pp. 1005-1016, June 2018.
  4. L. Sloot et al., “O 089 – A soft robotic exosuit assisting the paretic ankle in patients’ post-stroke: Effect on muscle activation during over ground walking”, Gait & Posture, 2018. DOI: 10.1016/j.gaitpost.2018.06.124
  5. L. Zhang, J. Li, M. Dong, Y. Cui and X. Rong, “Development of a Compatible Exoskeleton (Co-Exos II) for Upper-Limb Rehabilitation*,” 2019 16th International Conference on Ubiquitous Robots (UR), Jeju, Korea (South), 2019, pp. 240-245. DOI: 10.1109/URAI.2019.8768641
  6. V. P. G., S. Jacob, V. G. Menon, S. Rajesh and M. R. Khosravi, “Brain-Controlled Adaptive Lower Limb Exoskeleton for Rehabilitation of Post-Stroke Paralyzed,” in IEEE Access, 2019. DOI: 10.1109/ACCESS.2019.2921375
  7. Available online in http://www.cpsa.ca/wp-content/uploads/2015/03/EEG-Standards-April-2016-v21.pdf [Accessed on March 17, 2019]

Sunil Jacob is currently the Director, Centre for Robotics, SCMS School of Engineering and Technology and also Professor in Department of Electronics and Communication Engineering. He has been the recipient of Young Gandhian Technological Innovation Appreciation Award 2018 and AICTE Chhatra Vishwakarma Award 2017 in Electronics. His project Muscles to Machine Interface for Paralyzed has been funded by IEEE EPICS USA. Other funded projects include Bionic Haptic Arm, Rejuvenating the cells of human body, De-addiction coil for drug addicts, Smart Keyboard for Disabled Person, Low Cost 3D printer, Wearable device for detection and Prevention of Heart Failure, Bug-bot for Mosquito Attractor.

Dr. Varun G Menon is currently an Associate Professor in Department of Computer Science and Engineering, SCMS School of Engineering and Technology, India. He is an ACM Distinguished Speaker and has been the resource person for numerous workshops on “Quality Research Publishing” in various Engineering Colleges in India, primarily aimed at enhancing the quality of research publishing among the faculty and students. He is currently serving as Guest Editor for the Journal of Supercomputing, Future Internet Journal and Springer Nature Applied Sciences journal. He is also serving in the Editorial and Editorial Review Boards of many journals including Journal of Organizational and End User Computing, International Journal of E-Health and Medical Communications, International Journal of Disaster Response and Emergency Management. He is also currently serving in the Review Boards of many high impact factor journals including IEEE Transactions on Vehicular Technology, IEEE Transactions on Communications, IEEE Communications Magazine, IEEE Access, Ad-Hoc Networks Journal (Elsevier), Computer Communications Journal (Elsevier), Vehicular Communications Journal (Elsevier). His research interests include Internet of Things, Fog Computing and Networking, Underwater Acoustic Sensor Networks, Information Science, Scientometrics, Informatics of Scientific Databases, Mobile Ad-Hoc Networks, Wireless Communication, Opportunistic Routing, Wireless Sensor Networks, Educational Psychology, Cyber Psychology.

Dr. Saira Joseph received her M-Tech and PhD degree from Cochin University of Science and Technology, Kerala, India in 2006 and 2017 respectively. She is currently an Associate Professor in the Department of Electronics and Communication Engineering at SCMS School of Engineering and Technology, Kerala, India.  Her interests include Fractal antennas, UWB, RADAR and Metamaterials.


Mubashir Husain Rehmani (M’14-SM’15) received the B.Eng. degree in computer systems engineering from Mehran University of Engineering and Technology, Jamshoro, Pakistan, in 2004, the M.S. degree from the University of Paris XI, Paris, France, in 2008, and the Ph.D. degree from the University Pierre and Marie Curie, Paris, in 2011. He is currently an Assistant Professor at COMSATS Institute of Information Technology, Wah Cantt., Pakistan. He was a Postdoctoral Fellow at the University of Paris Est, France, in 2012. His current research interests include cognitive radio ad hoc networks, smart grid, wireless sensor networks, and mobile ad hoc networks. Dr. Rehmani served in the TPC for IEEE ICC 2016, IEEE GlobeCom 2016, CROWNCOM 2016, IEEE VTC Spring 2016, IEEE ICC 2015, IEEE WoWMoM 2014, IEEE ICC 2014, ACM CoNEXT Student Workshop 2013, IEEE ICC 2013, and IEEE IWCMC 2013 conferences. He is currently an Editor of the IEEE Communications Surveys and Tutorials and an Associate Editor of the IEEE Communications Magazine, IEEE Access journal, Elsevier Computers and Electrical Engineering (CAEE) journal, Elsevier Journal of Network and Computer Applications (JNCA), Ad Hoc Sensor Wireless Networks (AHSWN) journal, Springer Wireless Networks Journal, KSII Transactions on Internet and Information Systems, and the Journal of Communications and Networks (JCN). He is also serving as a Guest Editor of Elsevier Ad Hoc Networks journal, Elsevier Future Generation Computer Systems journal, IEEE Access journal, the IEEE Transactions on Industrial Informatics, Elsevier Pervasive and Mobile Computing journal and Elsevier Computers and Electrical Engineering journal. He has authored/ edited two books published by IGI Global, USA, one book published by CRC Press, USA, and one book is in progress with Wiley, U.K. He is the founding member of IEEE Special Interest Group (SIG) on Green and Sustainable Networking and Computing with Cognition and Cooperation. He received “Best Researcher of the Year 2015 of COMSATS Wah” award in 2015. He received the certificate of appreciation, “Exemplary Editor of the IEEE Communications Surveys and Tutorials for the year 2015” from the IEEE Communications Society. He received Best Paper Award from IEEE ComSoc Technical Committee on Communications Systems Integration and Modeling (CSIM), 2017.