BCI FOR PARALYSES PATIENT CONVERTING AUDIO TO VIDEO

BRAIN COMPUTER INTERFACE

1. “Design and Implementation of Patient Monitoring System Using Steady State Visual Evoked Potential Signal(SSVEP) Based on Labview” by Harathi Devi. Nalla

2. Agenda Introduction Literature Review Aim , Objective , Methodology Block Diagram Implementation of design Results and Discussion Conclusion Future Scope Project Schedule References

3. INTRODUCTION Brain-Computer Interface A brain-computer interface(BCI) is a system that allows a user to communicate his intent to a system without using any peripheral output pathways. Types of BCIs BCIs can be based on different neuromechanisms. - Sensorimotor activity - oddball paradigm - Visual evoked potentials(VEP)

4. BCIs based on the steady state-visual evoked potential appear to be promising BCI Neuromechanisms 10% 45% 30%

5. Literature Review Speech generation using Electromyography(EMG) techniques EMG recording from the vocal track movements are useful for the physically healthy individual but failed for the tetraplegic individuals who do not have an accurate voluntary control over the speech[3]. This technique is not feasible for paralysed individuals those who are suffering from lock-in syndrome[3]. The signal potential of EEG is more accurate compared to EMG[2]. The randomness of the EMG signal is stronger than that of the EEG signal, so the entropy stronger of EMG signals will be larger than that of the EEG signals[1].

6. Electrodes The EEG was measured with six channels at O1, Oz, O2, P3, Pz and P4, referenced to Fz and grounded at linked A1-A2, but only O1, Oz and O2 channels were used for on-line feedback. These positions were chosen since they are over visual cortex where SSVEP have higher amplitude[5,7]. Dry electrodes are equipped with six probes of spring loaded type. These probes contract their length when compressed. This structure provides flexibility and geometric adaption between the sensor and the irregular scalp surface[4]. The impedance can be obtained by measuring the voltage difference between a reference electrode and a target electrode[6].

7. Electroencephalography (EEG) The most practical, non-invasive, and non-expensive signal acquisition device of BCI systems is Electroencephalograms (EEGs) whereas Magnetoencephalogram (MEG) and Electrocorticogram (ECoG) are applied to a few BCI systems[4,9]. The EEG signals is a brain activity which is detected by electrodes placed on the scalp as a small electric signals for future processing this signal is first amplified and filtered by LPF followed by HPF[13,10]. A differential amplifier can be used to remove common mode interference signal such as noise[11,8]. When the subject focuses his attention on the stimulus, EEG activity is detected at the corresponding frequency over occipital areas[12,9]. The SSVEP-based BCI has many advantages over other EEG-based BCI systems, including a high signal-to-noise ratio[13].

8. Aim: To design and implement patient monitoring system which helps in enabling communication between a disabled person and his care taker using brain computer interfacing.

9. Objectives: To study the existing brain control devices and their limitations. To design and fabricate filters for signal processing. Locating and placing the 3disc electrode on the visual cortex part of the Brain. To design and implement a fixed frequency(25HZ) visual stimulation system. To construct a program in labview to analyze and implement wavelet transform. Designing and programming a system to convert processed signals into speech

10. Methodology Methodology for objective 1: Literature survey will be made to study the different brain computer interfacing systems available, and different technologies and limitations in those technologies. Different latencies available for visual evoked potentials and their limitations will be studied, and the best suitable latency will be used. Methodology for objective 2: The SSVEP (Steady State Visual Evoked Potential System) signals extracted from human subjected needs to be amplified. Various filters such as Band pass and notch filter are designed and fabricated to remove unwanted signal frequency. High gain amplifier using INA217 is designed and constructed to strengthen the weak SSVEP signals.

11. Methodology for objective 3: 10-20 international electrode system will be studied and specific points will be located on the visual cortex of the brain to place the 3 disc electrodes. A conductive gel will be used at the point of placement of electrodes and the conduction will be tested . Methodology for objective 4: Fixed frequency visual stimulation system is designed and implemented using a self designed computer application developed using HTML, CSS and JAVA Script. A tracing paper is placed on the light source of the stimulation system to reduce the intensity in order to suit human eye.

12. Methodology for objective 5: Based on the signal obtained from the human subject, a program is written in labview to analyze wavelet transform and power spectral density of signal is subtracted to define the amplitude of the signal at 25HZ. Methodology for objective 6: The signals from labview is used by a computer application to convert digital signal into speech.

13. PROPOSED BLOCK DIAGRAM : Visual stimulation SSVEP signals to Amplifier Filters(LPF,HPF& NOTCH) USB DAQ6009 LABVIEW Audio/text Power Spectral Density Computer Application Audio Non-inverting Amplifier

14. Implementation of design Circuit layout design using software express PCB. Toner transfer method is used to transfer toner to a copper clad. Amplifier

15. Clad with toner imprinted is treated with ferric chloride solution to get the required circuit design on it. PCB after the completion of etching process

16. Final circuit after soldering of components. Drilling of holes on PCB for component placement.

17. Screenshot of stimulation system with predefined messages

18. Acquiring of EOC signals to check the correctness of amplifier and filter circuits

19. EOC signal waveforms

20. When both input-1 and input-2 to the instrumentation amplifier INA217 are equal When input-1 and input-2 are of same frequency but different amplitudes Results and Discussion Results from phase1: amplifier testing

21. When input-1 and input-2 are of different frequency and different amplitude

22. Filter circuit design specification Lower cut-off frequency: 0.05Hz Higher cut-off frequency: 30Hz Bandstop frequency: 50Hz

23. Filter circuit output at 30Hz. Amplitude reduced to 500mVpp.

24. Filter circuit output at 35Hz. Amplitude reduced to 100mVpp.

25. Filter circuit output at 15Hz. Amplitude reduced to 2.93Vpp.

26. At 50Hz frequency, output is almost equal to 0V. F requency from AC source is thus eliminated

27. Block Diagram in LABVIEW

28. Front Panel in LABVIEW

29. The system was tested for a single frequency of 15Hz. Observed maximum peak amplitude from power spectral density.

30. Conclusion Simple power spectrum in Labview software(simpler method than the existing methods) could be used for feature extraction and finding the threshold for SSVEP. Flickering light on the black background contributed more potential than a lighter background. White light contributed maximum amplitude followed by red, green and blue color light.

31. Future Scope System which can accommodate more number of frequencies so that more messages can be embedded. System can be made more faster and accurate by experimenting with different light sources, intensities, patterns and frequencies. It can be made more portable and less dependent on high end software.

32. References: [1] YanyanZang and Gang Wang, ChaolinTeng , "Removal of EMG artifact from EEG signal by the multivariate empherierl made decomposition," International Journal of IEEE , pp. 875-876, Jan 2014. [2] Yong- Sheng Francis H. Y. Chan, "Fuzzy EMG classification for Prosthesis control," IEEE transactions on rehabilitation engineering , vol. 8, pp. 305-311, Sep 2012. [3] Toshio Tsuji Osamu Fukuda, "A Human-Assisting Manipulator Teleoperated by EMG signals and Arm," IEEE transactions on Robotics and , vol. 19, Apr 2015. [4] ValiUddin , UzmanNaz , SadiaParveen , Tariq Javid and Abdul MujeebMemon AzharDilshad , "On the Development of Novel, plug and play SSVEP – EEG based general purpose human-computer interaction device," Asian Journal of Engineering Sciences and Technology , pp. 1-3, 2016. [5] LaakkoMalmivw and Sari Ahokas , "High resolution EEG recording system using smart electrodes," International Journal of IEEE , pp. 21-24, March 2014. [6] Younghak shin Saungchan lee, "Dry electrode design and performance evaluation for EEG based Brain Computer," International Journal of IEEE , pp. 52-55, July 2013. [7] Stephen. R. Deiss , Tzyy -Ping Jung and Gert Cauwenberghs Thomas. J. Sullival , "A Brain–Machine Interface using Dry-Contact, Low-Noise EEG sensors," IEEE , pp. 1986-1989, 2008. [8] Resalat , FardinAfdideh SeyedNavid , "Real Time Monitor of Military Sentinel Sleepiness using a Novel SSVEP-Based," IEEE IMBS International conference of Biomedical engineering and sciences .

33. [9] V. S. Jabode , IEEE P. M. Shende , "Literature review of Brain Computer Interface (BCI) using EEG signal," IEEE , vol. 8-10, pp. 1435-1467, 2015. [10] D Habbal , Z Lugo, M Lebeau , P Horki , E Amico , C Pokorny D Lesenfants , "An independent SSVEP-based BCI in locked- in syndrome," Journal of Neural Engineering , vol. 1, pp. 223-226, 2014. [11] Chi and Gert Cauwenberghs Yum, " Micropower Non-contact EEG electrode with active common-mode noise suppression and input capacitance cancellation," 31st annual International Conference of the IEEE , pp. 4218-4221, Sep 2009. [12] Johannes Muller- Gerking and Gert Pfurtescheller Herbert Ramoser , "Optimal Spacial Filtering of single Trial EEG During Imagined Hand Movement," IEEE Transactionson Rehabilitation engineering , vol. 8, pp. 441-446, Dec 2000. [13] Asynchronous BCI control using high frequency SSVEP, "Pablo F Diez , Vicentre A," Journal of Neuro Engineering and Rehabilitation , 2011.  

34. THANK YOU