Downloaded 20 times
Uploaded byIsuru Jayarathne
Introduction to Brain-Computer Interface (BCI)
This document discusses brain-computer interfaces (BCI), covering applications like controlling robots and drones, as well as educational uses and emotion classification. It details the technologies involved, such as EEG and fMRI, the basics of EEG including signal types and electrode placement, and methods for signal processing and machine learning. Finally, it highlights the impact of biosignals and additional interesting technologies related to BCI.
More Related Content
Similar to Introduction to Brain-Computer Interface (BCI)
![[IJET-V1I5P6] Authors:M. Newlin Rajkumar, M. Lavanya](https://cdn.slidesharecdn.com/ss_thumbnails/ijet-v1i5p6-151022011423-lva1-app6891-thumbnail.jpg?width=600ounds&width=560&fit=bounds)
Introduction to Brain-Computer Interface (BCI)
- 1. Brain Computer Interface Isuru Jayarathne University ofAizu 10/01/2018
- 2. Content ◎ BCI relatedapplications ◎ Biosignals ◎ What is BCI ◎ Related subject areas ◎ Technologies ◎ Basics of EEG ◎ Brain lobes ◎ Electrode placement 2 ◎ Frequency bands ◎ Basic flow ◎ EEG devices ◎ Artifacts ◎ Pre-Processing ◎ Feature extraction ◎ Pattern recognition ◎ More...
- 3. BCI Applications ◎ Controllingrobots 3
- 4. BCI Applications ◎ Controllingdrones 4
- 5. BCI Applications ◎ BCIwith VR 5
- 6. BCI Applications ◎ BCIspeller 6
- 7. BCI Applications ◎ BCImouse 7
- 8. BCI Applications ◎ BCIexoskeleton 8
- 9.
- 10. BCI Applications ◎ Brainsignal-based authentication system ◎ Epilepsy seizure detection ◎ Meditation applications ◎ Educational applications ◎ Emotion classification ◎ Sleep disorder studies ◎ etc. 10
- 11. Biosignals ◎ Signals thatcan be continually measured and monitored from the living being ◎ Can be electrical or non-electrical ○ Electrocardiogram (ECG) ○ Electroencephalogram (EEG) ○ Magnetoencephalogram (MEG) ○ Electromyogram (EMG) ○ Electrooculography (EOG) ○ Galvanic skin response (GSR) ○ Mechanomyogram (MMG) 11
- 12. What is BCI ◎Direct communication between enhanced brain and an external device. ◎ Methods ○ Invasive ○ Partially invasive ○ Non-invasive 12
- 13. Related subject areas ◎Neuroscience ◎ Digital Signal Processing (DSP) ◎ Digital Image Processing (DIP) ◎ Machine Learning (ML) - in the most cases 13
- 14. Technologies ◎ fMRI: functionalMagnetic Resonance Imaging ◎ EEG: Electroencephalography ◎ MEG: Magnetoencephalography ◎ NIRS: Near-infrared spectroscopy 14
- 15. Basics of EEG ◎Electroencephalography (EEG): Electrophysiological monitoring method to record electrical activity of one’s brain ◎ Subtle electric field is generated by neurons ◎ EEG electrodes can capture the electric field of group of neurons 15
- 16. Brain lobes ◎ Fourbrain lobes ○ Frontal ○ Parietal ○ Temporal ○ Occipital 16
- 17. Electrode placement ◎ 10-20system ◎ 10-10 system or 10% ◎ 10-5 system 17
- 18. Electrode placement (Cont.) ◎10-20 system & 10% system 18
- 19.
- 20. EEG devices ◎ Clinical-and consumer-grade devices are available ◎ Two types of electrodes, Wet and dry ◎ Comes with different number of electrodes 20
- 21. EEG devices (Cont.) 21 DeviceNumber of Channels Electrode type NeuroSky Mindwave 1 Dry Muse 2 Dry EMOTIV Insight 5 Dry EMOTIV Epoc+ 14 Wet OpenBCI 16 Dry mBrainTrain 24 Wet ENOBIO 32 32 Dry Cognionics Mobile- 72 64 Dry
- 22. Pre-Processing ◎ Filtering ○ Separatefrequency bands ◎ Artefact removal ○ Blind Source Separation (BSS) ◎ Fourier Transform/Fast Fourier Transform 22
- 23. Frequency bands ◎ Delta:1 - 4 Hz ○ Highest amplitude ○ Presents during deep sleep ◎ Theta: 4 - 8 Hz ○ Focused attention, memory recall (frontal) ◎ Alpha: 8 - 13 Hz ○ Mental and physical relaxation (occipital, parietal) 23
- 24. Frequency bands (Cont.) ◎Beta: 13 - 30 Hz ○ Active thinking, active concentration ◎ Gamma: < 30 Hz ○ Binding of different populations of neurons 24
- 25. Artifacts ◎ EOG, EMG,ECG interferes the EEG signals ◎ Impact of power line noise ◎ Head swaying and swinging To remove ◎ Record other biosignals as well ◎ Filtering ◎ Record gyroscope data 25
- 26. Feature extraction ◎ PowerSpectral Density (PSD) ◎ Common Spatial Patterns (CSP) ◎ Autoregressive Coefficients (AR) ◎ Frontal Asymmetry ◎ Inter-Hemispheric Power Difference ◎ Inter-Hemispheric Linear Complexity ◎ Spectral Coherence ◎ Peak frequency 26
- 27. Pattern recognition/Machine Learning ◎Linear Discriminant Analysis ◎ Quadratic Discriminant Analysis ◎ Support Vector Machine (Linear, Quadratic, Cubic, Gaussian) ◎ k-Nearest Neighbour ◎ Deep Learning techniques 27
- 28. References ◎ IMOTION EEGPocket Guide (url: https://imotions.com/eeg-guide-ebook/) ◎ Wikipedia ◎ https://www.youtube.com/watch?v=Zd9WhJPa2Ok ◎ https://youtu.be/hLjxMjBlB9k ◎ https://www.youtube.com/watch?v=yamA5yM2h0Q ◎ https://www.youtube.com/watch?time_continue=87&v=tc82Z_yfEwc ◎ https://youtu.be/wKDimrzvwYA ◎ https://youtu.be/LvJk420Bbts ◎ https://youtu.be/jeLghZ8GASA ◎ www.emotiv.com ◎ www.choosemuse.com 28
- 29. Other interesting stuffs ◎rTMS, tDCS ◎ Neuromorphic chips ◎ Quantum Computers (ibm.biz/qx-introduction) ◎ Augmented Reality (AR) and Virtual Reality (VR) with HMDs and smart glasses 29
- 30.