![Adaptive Solutions
• How can we design solutions that more adaptive to the individual
characteristics of the user?
• BCIs is limited by the ability of users to provide distinguishable changes in
their neurophysiological input, also known as BCI literacy.
o Various factors affect this literacy and range from the person’s current fatigue level to physiological
makeup.
• ITF(individual technology fit) can be reflected by the individual’s
performance with the BCI technology.
• A methodology that ties performance to available BCI technologies
based on individual characteristics can greatly expedite the technology-
fit process.
• The extent that technology functionality matches task requirements and
individual abilities [and] is presumed to lead to higher performance.](https://image.slidesharecdn.com/ed88ded0-9632-4cb7-a93c-9cd81f54cdd2-160406011315/85/asjadpresentation-10-320.jpg)
asjadpresentation
- 1. Brain Computer Interfaces Muhammad Asjad
- 2. What is BCI?
- 3. BCI - Definition • “A system which takes a bio signal measured from a person and predicts(in real time / on a single-trial basis) some abstract aspect of the person’s cognitive state.” • Types: o Active BCI o Reactive BCI o Passive BCI
- 4. Brain Activity 10 secs of brain activity
- 5. Signals and Sensors • Invasive vs non – invasive. o Invasive techniques yield better results o invasive techniques are not usually used by healthy patients • How to measure brain activity? o EEG (Electroencephalogram) o Functional Near –infrared Spectrography (fNIRS) o Galvanic skin response (GSR)
- 6. Signals and Sensors Figure: Brain data visualization ECG headset
- 7. BCI is an interdisciplinary problem and borrows from: • Signal Processing • Machine Learning and statics • Computational Intelligence • Neuroscience • Cognitive Science • Problems are similar to Computer Vision, Speech Recognition, Pattern Recognition, Time- Series Analysis, Control System & Robotics
- 8. Applications • Communication for seriously disabled patients (e.g Tertraplegia, Locked-in Syndrome) • Prosthetic Control • Entertainment(gaming, thought control) • Home Automation • Operator monitoring (braking intent, fatigue,forensices etc.) • Neuroehabilitation
- 9. Challenges • Zero false activation rate is difficult to achieve. • Limited Transfer rates. E.g 10-25bits/min o neuroprosthesis control, may require higher information transfer rates • High Signal to noise ratio o Sensitive measures are hard to obtain. o Background activity can interfere with relevant brain activity. o Sophisticated signal processing is required • Sensor Placement. o Calibration is required. • Brain workings are complicated and still not fully understood. o Neurons are involved in multiple activities at once. • Processing depends on unkown parameters (person-specific, task-specific, otherwise variable). o Cortex is folded differently for every person. o Universal models don’t exist. i.e functional map is different for individuals. o Each individual can show different brain activity for similar tasks
- 10. Adaptive Solutions • How can we design solutions that more adaptive to the individual characteristics of the user? • BCIs is limited by the ability of users to provide distinguishable changes in their neurophysiological input, also known as BCI literacy. o Various factors affect this literacy and range from the person’s current fatigue level to physiological makeup. • ITF(individual technology fit) can be reflected by the individual’s performance with the BCI technology. • A methodology that ties performance to available BCI technologies based on individual characteristics can greatly expedite the technology- fit process. • The extent that technology functionality matches task requirements and individual abilities [and] is presumed to lead to higher performance.
- 11. Adaptive Solutions cont. • identify salient individual user characteristics that matches with features of brain- computer interface technologies • Individuals vary in their characteristics across many dimensions o Characteristics are a person’s demographic, physiological, and cognitive traits. • Determine effective approaches, paradigms and heuristics that link individual characteristics to available technologies.
- 12. ITF and Technology features • individual-technology fit is the extent to which individual characteristics match with technology features to enable a person’s control of a technology. • Technology Characteristics o Type – Classification of the general mechanism used o Biorecording Technology - Approach used to record signals from the participants (e.g., EEG, fNIR, fMRI, GSR). o Inputs – Placement of sensors/electrodes (e.g., areas over the brain, fingers) o Neurological Phenomenon – Phenomenon used to control the transducer (i.e., phenomena in electrical brain activity, phenomena in blood oxygenation, or phenomena in skin conductance). o Feature Extraction/Translation Algorithms – Component that extracts and translates the signal into a useful control signal
- 13. Individual Characteristics - Existing Studies • Little is known about which individual characteristics best match with particular BCI technologies • Research by Randolph and Moore Jackson: o proposed a set of characteristics affecting BCI technology control and tested them with fNIR- and GSR-based technologies o age, regular consumption of caffeine, and years of education all positively correlated with fNIR control o age, sex, hair and skin color, hair texture, meditation, regular consumption of alcohol, and video game experience all positively correlated with GSR control o interaction of age and hand- and-arm movement predicted modulation of the mu rhythm in a mu-based BCI
- 14. Research by Adriane B. Randolph (Kennesaw State University )• Focused on the mu rhythm(a type of brain signal), is based on continuous electrical variations in the motor cortex region of the brain according to real and imagined movement. • Mu-based BCIs can take advantage of the difference in signal properties between idle and active imagery within the motor cortex region of the brain to produce a control signal. • The proportional difference in signal properties is measured by a response R-squared value and indicates signal strength or the degree of modulation a person may induce • Results: instrument playing, being on affective drugs, sex, and age also play a key role in predicting mu rhythm modulation
- 15. Conclusion • More research should be done to further understand the differences between individuals, technology, and the impacts on BCI design. • This will allow assistive technology practitioners to better incorporate information about their users and refine their design efforts.
- 16. References Ebrahimi, T., "Recent advances in brain-computer interfaces," Multimedia Signal Processing, 2007. MMSP 2007. IEEE 9th Workshop on , vol., no., pp.17,17, 1-3 Oct. 2007 doi: 10.1109/MMSP.2007.4412807 Zander TO, Lehne M, Ihme K, Jatzev S, Correia J, Kothe C, Picht B, Nijboer, F A dry EEG-system for scientific research and brain-computer interfaces Frontiers in Neuroprosthetics, in press. Randolph, A.B., "Not All Created Equal: Individual-Technology Fit of Brain-Computer Interfaces," System Science (HICSS), 2012 45th Hawaii International Conference on , vol., no., pp.572,578, 4-7 Jan. 2012 Ebrahimi, T., "Recent advances in brain-computer interfaces," Multimedia Signal Processing, 2007. MMSP 2007. IEEE 9th Workshop on , vol., no., pp.17,17, 1-3 Oct. 2007 Lightbody, G.; Ware, M.; McCullagh, P.; Mulvenna, M.D.; Thomson, E.; Martin, S.; Todd, D.; Medina, V.C.; Martinez, S.C., "A user centred approach for developing Brain-Computer Interfaces," Pervasive Computing Technologies for Healthcare (PervasiveHealth), 2010 4th International Conference on- NO PERMISSIONS , vol., no., pp.1,8, 22-25 March 2010
- 17. References cont. • Das, K.; Rizzuto, D.S.; Nenadic, Z., "Mental State Estimation for Brain--Computer Interfaces," Biomedical Engineering, IEEE Transactions on , vol.56, no.8, pp.2114,2122, Aug. 2009 • • Kottaimalai, R.; Rajasekaran, M.P.; Selvam, V.; Kannapiran, B., "EEG signal classification using Principal Component Analysis with Neural Network in Brain Computer Interface applications," Emerging Trends in Computing, Communication and Nanotechnology (ICE-CCN), 2013 International Conference on , vol., no., pp.227,231, 25-26 March 2013 • Faradji, F.; Ward, R.K.; Birch, G.E., "A brain-computer interface based on mental tasks with a zero false activation rate," Neural Engineering, 2009. NER '09. 4th International IEEE/EMBS Conference on , vol., no., pp.355,358, April 29 2009-May 2 2009 • Heung-Il Suk; Seong-Whan Lee, "A Novel Bayesian Framework for Discriminative Feature Extraction in Brain-Computer Interfaces," Pattern Analysis and Machine Intelligence, IEEE Transactions on , vol.35, no.2, pp.286,299, Feb. 2013 • Mamun, K.A.; Huda, M.N.; Mace, M.; Lutman, M.E.; Stein, J.; Liu, X.; Aziz, T.; Vaidyanathan, R.; Wang, S., "Pattern classification of deep brain local field potentials for brain computer interfaces," Computer and Information Technology (ICCIT), 2012 15th International Conference on , vol., no., pp.518,523, 22-24 Dec. 2012 • McCormick, M.; Rui Ma; Coleman, T.P., "An analytic spatial filter and a hidden Markov model for enhanced information transfer rate in EEG-based brain computer interfaces," Acoustics Speech and Signal Processing (ICASSP), 2010 IEEE International Conference on , vol., no., pp.602,605, 14-19 March 2010 • Xie, S.Q.; Gao, C.; Yang, Z.L.; Wang, R.Y., "Computer-brain interface," Neural Interface and Control, 2005. Proceedings. 2005 First International Conference on , vol., no., pp.32,36, 26-28 May 2005