Research a Collection of Research Reviews on Rehabilitation Topics from NARIC and Other Information Resources

reSearch A collection of research reviews on rehabilitation topics from NARIC and other information resources.

Volume 10, Issue 2, 2015 Brain Computer Interface & Neuromuscular Diseases/Disorders n this edition of reSearch we explore the topic IEEE Explore, and PubMed databases can be found of brain computer interface (BCI) technology at the end of this document. Ifor individuals with neuromuscular diseases/ A search of the REHABDATA database resulted disorders. Brain computer interface (BCI), also in 18 documents published between 1995 and 2015. known as brain- or mind-machine interface, is a direct The CIRRIE and ERIC database searches resulted in 2 communication interface system between an external documents between 1996 and 2002, and 2 documents device and the brain, which allows an individual from 1993 and 2013; respectively. The Cochrane to communicate with, or control a computer or database searches resulted in 2 documents from 2004 other electronic device, using his or her brainwaves and 2014. Finally, a search of the IEEE Explore and without movement from the neuromuscular system. PubMed databases resulted in 23 documents between (Retrieved from: http://www.brainvision.co.uk/ 1998 and 2014, and 33 documents between 2002 blog/2014/04/the-brief-history-of-brain-computer- and 2015; respectively. The complete citations are interfaces). There are two types of BCI systems: included in this research brief. invasive and noninvasive. Invasive systems require a surgical procedure to implant electrodes on or near References: the surface of the brain; whereas, noninvasive systems involve the use of a cap (similar in appearance to a Fried-Oken, M. & Peters, B. (2014). Brain-computer fabric swimming cap) that keep applied electrodes interface (BCI). ALS Association. Retrieved from in place on the scalp. The noninvasive systems http://www.alsa.org/als-care/resources/publications- require the use of a conductive gel and provide little videos/factsheets/brain-computer-interface.html. or no discomfort other than needing to wash the gel out of the hair after use. Whether the BCI system is (2014, April 30). The brief history of brain computer invasive or noninvasive, the process is the same—the interfaces. Brain Vision UK Blog. Retrieved from brain signals picked up by the electrodes are sent to http://www.brainvision.co.uk/blog/2014/04/the-brief- the computer and software translates these signals history-of-brain-computer-interfaces. into computer commands. (Retrieved from: http:// www.alsa.org/als-care/resources/publications-videos/ factsheets/brain-computer-interface.html). Table of Contents This edition of reSearch provides a “snapshot” of over 20 years of research on BCI. This “snapshot” NIDILRR Funded Projects ...... 2 page presents a general overview of the BCI and its use NARIC Citations ...... 3 page with persons with neuromuscular diseases/disorders CIRRIE Citations ...... 10 page (i.e., amyotrophic lateral sclerosis [ALS], locked- ERIC Citations ...... 10 page in syndrome [LIS], stiff person syndrome [SPS], Cochrane Citations ...... 11 page multiple sclerosis [MS], etc.). The combined search IEEE Xplore Citations...... 12 page terms for this edition of reSearch included: brain PubMed Citations ...... 19 page computer interface, BCI, neuromuscular diseases/ Quick Looks ...... 32 page disorders, ALS, MS, locked-in, stiff person, and motor neuron. A listing of over 200 additional descriptor Search Terms ...... 34 page terms between the NARIC, CIRRIE, ERIC, Cochrane, Page 2 NIDILRR Funded Projects Related to University of Maryland Advanced Neuromotor Brain Computer Interface & Rehabilitation Research Training (UMANRRT) Neuromuscular Diseases/Disorders Project Number: H133P100014 Phone: 410/706-1771 In addition to document searches, we searched our Email: [email protected]. NIDILRR Program Database to locate grantees/projects related to BCI & NMD. The search resulted in These projects have completed their research activi- six currently funded and six projects that are no longer ties and are now closed. active. Project information and their publications are offered as additional resources for our patrons. An Adaptive Haptic Interface for Individuals with Disabilities Advanced Rehabilitation Research Training Cen- Project Number: H133S040119 ter on Neuro-musculoskeletal Rehabilitation Phone: 434/973-1215 Project Number: H133P070007 Email: [email protected]. Phone: 973/324-3550 www.barron-associates.com. Email: [email protected]. kesslerfoundation.org/researchcenter/postdoctoralfel- Assistive Computer Interfaces for Persons with lowshipprogram.php. Movement Disorders Project Number: H133G30064 Advanced Rehabilitation Research Training in Phone: 410/955-7093 Neuromuscular and Neurodevelopmental Disor- Email: [email protected]. ders Project Number: H133P110005 Brain Computer Interface to Enable Improved Phone: 916/734-5292 Communication Access www.ucdmc.ucdavis.edu/pmr/education/fellowship. Project Number: H133S120039 html. Phone: 763/515-5355 Email: [email protected]. Development of an Intelligent Assistive Robotic System for Individuals with Multiple Sclerosis Personalized Synthetic Speech Using ModelTalk- Project Number: H133G120275 er: Development and Evaluation Phone: 407/882-2820 Project Number: H133G990182 Email: [email protected]. Phone: 302/651-6835 www.eecs.ucf.edu/~abehal. Email: [email protected]. www.asel.udel.edu/speech. Rehabilitation Engineering Research Center on Augmentative and Alternative Communication Preparations for In-Home Testing of Brain-Com- (The RERC on AAC) puter Interfaces Operating Assistive Technology Project Number: H133E140026 Project Number: H133G090005 Phone: 814/863-2010 Phone: 734/936-7170 Email: [email protected]. Email: [email protected]. rerc-aac.psu.edu. www.umich.edu/~umdbi.

Rehabilitation Research and Training Center in A Synergy-Based Brain Computer Interface to Neuromuscular Diseases (RRTC-NMD) Reanimate Paralyzed Hands Project Number: H133B090001 Project Number: H133F100001 Phone: 916/734-4280 Phone: 610/306-8539 Email: [email protected]. Email: [email protected]. www.ucdmc.ucdavis.edu/pmr/research/index.html. Page 3 Documents from NARIC’s Bacher, D., Cash, S.S., Friehs, G., Hochberg, L.R., REHABDATA search listed Jarosiewicz, B., Masse, N.Y., Newell, K., Oakley, are listed below: E.M., Simeral, J.D., & Stavisky, S.D. (2015). Neural point-and-click communication by a person with incomplete locked-in syndrome. Neurorehabilita- 2015 tion and Neural Repair, 29(5), 462-471. NARIC Accession Number: J71452 Aloise, F., Arico, P., Cincotti, F., Desideri, L., Holz, ABSTRACT: Study evaluated the ability of an indi- E.M., Kubler, A., Leotta, F., Mattia, D., Riccio, A., & vidual with incomplete locked-in syndrome enrolled Rimondini, M. (2015). Hybrid P300-based brain- in the BrainGate Neural Interface System pilot clini- computer interface to improve usability for people cal trial to communicate using neural point-and-click with severe motor disability: Electromyographic control. Neural interface systems, also called brain- signals for error correction during a spelling task. computer interfaces (BCIs), are devices that connect Archives of Physical Medicine and Rehabilitation, the nervous system to an external device for the 96(3, Supplement 1), S54-S61. purpose of restoring mobility and communication to NARIC Accession Number: J70942 individuals with paralysis and anarthria (inability to speak) resulting from neurological disorders. A gener- ABSTRACT: Study evaluated the impact of a hybrid al-purpose interface was developed to provide control control on usability of a P300-based brain-computer of a computer cursor in tandem with one of two on- interface (BCI) system that was designed to control screen virtual keyboards: (1) a standard QWERTY an assistive technology software and was integrated keyboard and (2) the BrainGate Radial Keyboard, with an electromyographic channel for error correc- which has been designed to improve typing perfor- tion. The 11 participants in this pilot study included mance and ease-of-use for neural point-and-click 8 healthy and 3 severely motor-impaired subjects. communication. The participant used this interface The three people with severe motor disability were to communicate face-to-face with research staff by identified as potential candidates to benefit from the using text-to-speech conversion, and remotely using proposed hybrid BCI system for communication an Internet chat application. The Radial Keyboard and environmental interaction. To investigate the was compared to the standard QWERTY keyboard in improvement in usability, two modalities of BCI a balanced copy-spelling task. Results showed that the Radial Keyboard yielded a significant improvement system control were compared: a P300-based and a in typing accuracy and speed, enabling typing rates hybrid P300 electromyographic-based mode of con- greater than 10 correct characters per minute. Fur- trol. System usability was evaluated according to the thermore, the participant reported strongly preferring following outcome measures within three domains: the Radial Keyboard over the QWERTY keyboard, (1) effectiveness (overall system accuracy and P300- citing its speed and ease of use. This study demon- based BCI accuracy); (2) efficiency (throughput time strates the first use of an intracortical BCI for neural and users’ workload); and (3) satisfaction (users’ point-and-click communication by an individual with satisfaction). The information transfer rate and time incomplete locked-in syndrome. for selection were also considered. Results obtained in healthy participants were in favor of a higher us- Bieker, G., Fried-Oken, M., Heckman, S.M., Hug- ability of the hybrid control as compared with the gins, J.E., Peters, B., Wolf, C., & Zeitlin, D. (2015). non-hybrid. A similar trend was indicated by the Brain-computer interface users speak up: The virtual users’ forum at the 2013 international observational results gathered from each of the three brain-computer interface meeting. Archives of potential end-users. The proposed hybrid BCI control Physical Medicine and Rehabilitation, 96(3, Supple- modality could provide end-users with severe motor ment 1), S33-S37. disability with an additional option to exploit some NARIC Accession Number: J70939 residual muscular activity. Project Number: H133G090005 ABSTRACT: Article reports on the Virtual Users’ Forum, including initial results from ongoing research Page 4 being conducted by two brain-computer interface level of exhaustion, satisfaction, frustration, and (BCI) groups, with a goal of encouraging widespread enjoyment. To monitor BCI home use, evaluation use of participatory action research in all BCI efforts. data of every session were automatically sent and More than 300 researchers gathered at the 2013 In- stored on a remote server. Satisfaction with the BCI ternational BCI Meeting to discuss current practice as an assistive device and subjective workload was and future goals for BCI research and development. indicated by the patient. Usability of the BCI was The authors organized the Virtual Users’ Forum at the evaluated in terms of its effectiveness, efficiency, and meeting to provide the BCI community with feedback satisfaction. The influence of the BCI on quality of from users. Online surveys and in-person interviews life of the end user was assessed. The patient painted were used to solicit feedback from people with dis- in about 200 BCI sessions (1 to 3 times per week) abilities who are expert and novice BCI users. Their with a mean painting duration of 81.86 minutes. In responses were organized into four major themes: cur- most of the BCI sessions, the end user’s satisfaction rent (non-BCI) communication methods, experiences was high. Dissatisfaction occurred mostly because of with BCI research, challenges of current BCIs, and technical problems at the beginning of the study or future BCI developments. Two authors with severe varying BCI control. The subjective workload was disabilities gave presentations during the Virtual Us- moderate. Brain Painting had a positive impact on the ers’ Forum, and their comments are integrated with patient’s competence, adaptability, and self-esteem. the other results. While participants’ hopes for BCIs Results demonstrate that independent home use of a of the future remain high, their comments about avail- BCI-controlled painting application is feasible and able systems mirror those made by consumers about positively influences quality of life and supports conventional assistive technology. They reflect con- social inclusion of the patient. cerns about reliability (e.g., typing accuracy/speed), utility (e.g., applications and the desire for real-time Calabrese, B., Caruso, M., Cincotti, F., Frasca, V., interactions), ease of use (e.g., portability and sys- Inghilleri, M., Liberati, G., Mattia, D., Mecella, M., tem setup), and support (e.g., technical support and Pizzimenti, A., Riccio, A., Schettini, F., & Simione, caregiver training). People with disabilities, as target L. (2015). Assistive device with conventional, users of BCI systems, can provide valuable feedback alternative, and brain-computer interface inputs and input on the development of BCI as an assistive to enhance interaction with the environment for technology. To this end, participatory action research people with amyotrophic lateral sclerosis: A fea- should be considered as a valuable methodology for sibility and usability study. Archives of Physical future BCI research. Medicine and Rehabilitation, 96(3, Supplement 1), S46-S53. Botrel, L., Holz, E.M., Kaufmann, T., & Kubler, A. NARIC Accession Number: J70941 (2015). Long-term independent brain-computer ABSTRACT: Study evaluated the feasibility and interface home use improves quality of life of a usability of an assistive technology (AT) prototype patient in the locked-in state: A case study. Ar- operated with a P300-based brain-computer inter- chives of Physical Medicine and Rehabilitation, 96(3, face (BCI) to provide users with different degrees Supplement 1) , S16-S26. of muscular impairment resulting from amyotrophic NARIC Accession Number: J70937 lateral sclerosis (ALS) with communication and en- ABSTRACT: This case study demonstrated inde- vironmental control applications. The functionalities pendent brain-computer interface (BCI) home use of the AT prototype were implemented to control by a 73-year-old patient with amyotrophic lateral electronic devices available at an apartment-like sclerosis in the locked-in state and the effect it has on space designed to be fully accessible by people with quality of life. The P300 BCI-controlled Brain Paint- motor disabilities for occupational therapy, placed in ing application was installed at the patient’s home. a neurologic rehabilitation hospital. The AT prototype Family and caregivers were trained in setting up the was evaluated by eight end-users with ALS using BCI system. After every BCI session, the end user three experimental conditions based on: (1) a widely indicated subjective level of control, loss of control, validated P300-based BCI alone; (2) the AT prototype Page 5 operated by a conventional/alternative input device and appropriate brain activation during the initial tailored to the specific end-user’s residual motor assessment, and real-time feedback was provided to abilities; and (3) the AT prototype accessed by a P300- improve arousal. Consistent activation was observed based BCI. These three conditions were presented in multiple sessions. The EEG-based assessment to all participants in three different sessions. System showed that patients in a minimally conscious state usability was evaluated in terms of effectiveness may have the capacity to operate a simple BCI-based (accuracy), efficiency (written symbol rate, time for communication system, even without any detectable correct selection, workload), and end-user satisfaction volitional control of movement. (overall satisfaction) domains. A comparison of the data collected in the three conditions was performed. Chiodo, A.E., Huggins, J.E., Moinuddin, A.A., & Effectiveness and end-user satisfaction did not signifi- Wren, P.A. (2015). What would brain-computer cantly differ among the three experimental conditions. interface users want: Opinions and priorities of Condition three was less efficient than condition two potential users with spinal cord injury. Archives of as expressed by the longer time for correct selection. Physical Medicine and Rehabilitation, 96(3, Supple- Results indicate that a BCI can be used as an input ment 1) , Pgs. S38-S45. channel to access an AT by persons with ALS, with NARIC Accession Number: J70940 no significant reduction of usability. Project Number(s): H133G090005; H133N110002 ABSTRACT: Study surveyed people with spinal cord Carroll, A., Coyle, D., McCreadie, K., McElligott, injury (SCI) to determine their priorities for brain- J., & Stow, J. (2015). Sensorimotor modulation computer interface (BCI) applications and design assessment and brain-computer interface training features along with the time investment and risk in disorders of consciousness. Archives of Physi- acceptable to obtain a BCI. Forty people with SCI cal Medicine and Rehabilitation, 96(3, Supplement participated, including of 30 individuals identified 1), S62-S70. from the SCI research registry who were surveyed NARIC Accession Number: J70943 by telephone and 10 participants from a BCI usage ABSTRACT: Study assessed awareness in four sub- study who were surveyed in person before BCI use. jects in a minimally conscious state after brain injury, Descriptive statistics of functional independence, using an electroencephalogram (EEG)-based brain- living situations and support structures, ratings of computer interface (BCI), and determined whether importance of different task and design features, these patients may learn to modulate sensorimotor and acceptable levels of performance, risk, and time rhythms with visual feedback, stereo auditory feed- investment were calculated from the survey data. back, or both. The initial EEG-based assessment in- Participants were classified as having low function or cluded imagined hand movement or toe wiggling and high function based on their Functional Independence was used to activate sensorimotor areas and modulate Measure motor scores. Results indicated that BCIs brain rhythms during 90 trials for each subject. With- were of interest to 96 percent of the low-function in-subject and within-group analyses were performed group. Emergency communication was the top pri- to evaluate significant activations. A within-subject ority task. The most important design features were analysis was performed involving multiple BCI tech- “functions the BCI provides” and “simplicity of BCI nology training sessions to improve the capacity of setup.” Desired performance was 90 percent accuracy, the user to modulate sensorimotor rhythms through with standby mode errors no more than once every 4 visual and auditory feedback. Awareness detection hours and speeds of more than 20 letters per minute. was determined from sensorimotor patterns that dif- Dry electrodes were preferred over gel or implanted fered for each motor imagery task. BCI performance electrodes. Median acceptable setup time was 10 to was determined from the mean classification accuracy 20 minutes, satisfying 65 percent of participants. of brain patterns by using a BCI signal processing Creating BCI functions appropriate to the needs of framework and assessment of performance in mul- those with SCI will be of ultimate importance for BCI tiple sessions. All subjects demonstrated significant acceptance with this population. Page 6 Daly, J.J., Huggins, J.E. (2015). Brain-computer PLIS were 100 percent accurate on all tasks; three interface: Current and emerging rehabilitation PLIS missed single items. The results indicate that applications. Archives of Physical Medicine and the RSVP BCI screening protocol is a brief, repeat- Rehabilitation, 96(3, Supplement 1), S1-S7. able technique to identify the presence/absence of NARIC Accession Number: J70935 skills for BCI use by patients with different levels of ABSTRACT: Article introduces a journal supple- locked-in syndrome. ment that reviews research on the development and capabilities of brain-computer interface (BCI) for 2014 rehabilitation. A BCI can be defined as a system that translates “brain signals into new kinds of outputs.” Erdogmus, D., Fowler, A., Fried-Oken, M.B., Miller, After brain signal acquisition, the BCI evaluates the M., Mooney, A., Oken, B.S., Orhan, U., Peters, B., & brain signal and extracts signal features that have Roark, B. (2014). Brain-computer interface with proven useful for task performance. The articles in language model-electroencephalography fusion this supplement provide examples of work conducted for locked-in syndrome. Neurorehabilitation and using a variety of BCI technology applications, in- Neural Repair, 28(4), 387-394. cluding communication, leisure activities, and motor NARIC Accession Number: J68504 learning. These articles arose from presentations at the ABSTRACT: Study developed the RSVP Keyboard 2013 International Brain-Computer Interface Meet- to address the communication needs of individuals ing, which was held June 3 through June 7, 2013, in with locked-in syndrome (LIS), using a noninvasive Pacific Grove, California. brain-computer interface (BCI) that involves rapid serial visual presentation (RSVP) of symbols and a Fried-Oken, M., Mooney, A., Oken, B., & Peters, B. unique classifier with electroencephalography (EEG) (2015). A clinical screening protocol for the RSVP and language model fusion. BCIs use brain signals keyboard brain-computer interface. Disability and to provide a nonmotor communication channel for Rehabilitation: Assistive Technology, 10(1), 11-18. people with severely limited motor control. The NARIC Accession Number: J71072 RSVP Keyboard is the first BCI device developed ABSTRACT: Article describes the development of a with several unique features for people with LIS. screening protocol that identifies the sensory, motor, Individual letters are presented at 2.5 per second. cognitive, and communication skills that people with Computer classification of letters as targets or non- locked-in syndrome (PLIS) require in order to use the targets based on EEG is performed using machine RSVP Keyboard™ brain-computer interface (BCI). A learning that incorporates a statistical language model multidisciplinary clinical team of seven individuals for letter prediction via Bayesian fusion enabling representing five disciplines identified the requisite targets to be presented only one to four times. After skills for use of the RSVP Keyboard™. They chose screening, subjects first calibrated the system, and questions and subtests from existing standardized then completed a series of balanced word generation instruments for auditory comprehension, reading and mastery tasks that were designed with five incremen- spelling, modified them to accommodate nonverbal tal levels of difficulty, which increased by selecting response modalities, and developed novel tasks to phrases for which the utility of the language model screen visual perception, sustained visual attention decreased naturally. Six participants with LIS and and working memory. Questions were included about nine healthy controls completed the experiment. All sensory skills, positioning, pain interference and LIS participants successfully mastered spelling at medications. The result is a compilation of questions, level 1 and one subject achieved level five. Six of adapted subtests and original tasks designed for this nine control participants achieved level 5. Findings new BCI system. It was administered to 12 PLIS and suggest that individuals with incomplete LIS may 6 healthy controls. Administration required one hour benefit from an EEG-based BCI system that relies on or less. Yes/no choices and eye gaze were adequate EEG classification and a statistical language model. response modes for PLIS. Healthy controls and nine Steps to further improve the system are discussed. Page 7 Gruis, K.L., Huggins, J.E., & Thompson, D.E. 9 age-matched controls performed the cognitive task (2014). A plug-and-play brain-computer inter- of operating a brain-computer interface while electro- face to operate commercial assistive technology. encephalography was recorded over three sessions. Disability and Rehabilitation: Assistive Technology, Using normalized symbolic transfer entropy, directed Volume 9(2), 144-150. functional connectivity was measured from frontal to NARIC Accession Number: J68305 parietal (feedback connectivity) and parietal to frontal Project Number: H133G090005 (feedforward connectivity) regions. Results showed ABSTRACT: Study assessed the performance impact that feedback connectivity was not significantly dif- of using a brain-computer interface (BCI) as a plug- ferent between groups, but feedforward connectivity and-play input device to operate commercial assistive was significantly higher in individuals with ALS. technology (AT). BCIs are intended to enable people This result was consistent across a broad electroen- with severe physical impairments to communicate and cephalographic spectrum, and in theta, alpha, and to operate technology without moving their bodies. beta frequency bands. Feedback connectivity has Thirty-three participants (11 with amyotrophic lateral been associated with conscious state and was found sclerosis and 22 controls) were asked to operate two to be independent of ALS symptom severity in this devices (a commercial communication aid and a study, which may have significant implications for separate laptop computer) using a BCI. Results were the detection of consciousness in individuals with compared to traditional BCI operation by the same advanced ALS. The authors suggest that increases in users. Performance was assessed using both accuracy feedforward connectivity represent a compensatory and BCI utility, a throughput metric. Ninety-five per- response to the ALS-related loss of input such that cent confidence bounds on performance differences sensory stimuli have sufficient strength to cross the were calculated using a linear mixed model. The ob- threshold necessary for conscious processing in the served differences in accuracy and throughput were global neuronal workspace. small and not statistically significant. The confidence bounds indicate that if there is a performance impact 2012 of using a BCI to control an AT device, the impact could easily be overcome by the benefits of the AT Andone, I., Anghelescu, A., Ciurea, A.V., Daia, C., device itself. The results suggest that BCI control Danoczy, M., Fazli, S., Grozea, C., Mihaescu, A.S., of AT devices is possible, and the performance dif- Mirea, A., Onose, G., Popescu, C., Popescu, F., ference appears to be very small. BCI designers are Sinescu, C. J., Spanu, A., & Spircu, T. (2012). On encouraged to incorporate standard outputs into their the feasibility of using motor imagery EEG-based design to enable future users to interface with familiar brain-computer interface in chronic tetraplegics AT devices. for assistive robotic arm control: A clinical test and long-term post-trial follow-up. Spinal Cord 2013 (formerly Paraplegia), 50(8), 599-608. NARIC Accession Number: J64436 Blain-Moraes, S., Huggins, J.E., Lee, H., Lee, U., & ABSTRACT: Study assessed the feasibility of us- Mashour, G.A. (2013). Altered cortical communi- ing an electroencephalography (EEG)-based brain- cation in amyotrophic lateral sclerosis. Neurosci- computer interface (BCI) system, including a robotic ence Letters, 543, 172-176. arm, for reaching/grasping assistance in patients with NARIC Accession Number: J66657 tetraplegia. The study involved three consecutive Project Number: H133G090005 phases: training/calibration, feedback, and robot con- ABSTRACT: Study explored whether amyotrophic trol. Nine patients underwent EEG-BCI preliminary lateral sclerosis (ALS), a disorder associated primarily training and robot control sessions. Statistics entailed with the degeneration of the motor system, is associ- multiple linear regressions and cluster analysis. A ated with alterations in frontoparietal communication questionnaire-based follow-up, including patients’ that can be assessed with electroencephalography dur- perception of their EEG-BCI control capacity, was ing a cognitive task. Eleven individuals with ALS and continued up to 14 months after initial experiments. Page 8 Results indicated that EEG-BCI training/calibration- 2011 phase classification accuracy averaged 81.0 percent. Feedback training sessions averaged 70.5 percent Andrade, A.O., Bourhis, G., Losson, E., Naves, accuracy for 7 subjects who completed at least one E.L.M., Pinheiro Jr., C.G., & Pino, P. (2011). Al- feedback training session. Seven (77.7 percent) of ternative communication systems for people with the 9 subjects reported having felt control of the cur- severe motor disabilities: A survey. Biomedical sor; and 3 subjects (33.3 percent) felt that they were Engineering Online, 10(31). also controlling the robot through their movement NARIC Accession Number: J66400 imagination. No significant side effects occurred. BCI Available in full-text at: http://www.biomedical- performance was positively correlated with beta EEG engineering-online.com/content/10/1/31. spectral power density. Another possible influence ABSTRACT: Article provides a review of the way was the American Spinal Injury Association Impair- three bioelectrical signals (electromyographic, elec- ment Scale sensory score; higher sensory deficit may trooculographic, and electroencephalographic) have mean a slightly increased ability to perform motor been utilized in alternative communication with patients suffering severe motor impairment. It also imagery-based BCI. offers a comparative study of the various methods applied to measure the performance of alternative and Blain-Moraes, S., Gruis, K.L., Huggins, J.E., Schaff, augmented communication (AAC) systems. Commu- R., & Wren, P.A. (2012). Barriers to and media- nication aid devices using electromyographic signals tors of brain-computer interface user acceptance: can be subdivided into three major groups: mouse Focus group findings. Ergonomics, 55(5), 516-525. emulation, speech recognition, and switch-based NARIC Accession Number: J63934 control device. Being a manifestation of eye move- Project Number(s): H133G090005; H133N110002; ments, electrooculographic signals are processed to H133P090008; H133P090013 identify gaze, usually for cursor control on the screen. ABSTRACT: Article explores human factor issues However, the most popular method for gaze estima- involved in designing and evaluating brain-computer tion is using infrared cameras, through the reflection interface (BCI) systems for users with severe motor in the eye structures and their geometric relations. For disabilities. BCIs are designed to enable individuals people with severe motor disability such as locked-in with severe motor impairments such as amyotrophic syndrome, it often becomes impossible to communi- lateral sclerosis (ALS) to communicate and control cate or control a muscular activity. However, these their environment. By translating signals recorded people generally keep cerebral and sensory functions directly from the brain into computer output or intact. A solution to overcome this is to use electro- environmental control, BCIs may enable a motion- encephalography associated with the cerebral activity less, voiceless individual to articulate autonomy and to control an interface. This type of interface using subjectivity, communicate and develop relationships, the cerebral waves is usually called brain-computer thereby continuing to affirm their personhood. A focus interface. The article concludes with a review of the group was conducted with 8 individuals with ALS methods reported in the literature aiming at measuring who were familiar with BCI technology and nine the performances of AAC systems with and without of their caregivers to determine the barriers to and the use of human-machine interaction models. mediators of acceptance of BCI. Two key categories emerged: personal factors and relational factors. Per- 2006 sonal factors, which included physical, physiological and psychological concerns, were less important to Branner, A., Caplan, A.H., Chen, D., Donoghue, J.P., participants than relational factors, which included Friehs, G.M., Hochberg, L.R., Mukand, J.A., Penn, corporeal, technological, and social relations with R.D., Saleh, M., & Serruya, M.D. (2006). Neuronal the BCI. The importance of these relational factors ensemble control of prosthetic devices by a human was analyzed with respect to published literature on with tetraplegia. Nature, 442(7099), 164-171. actor-network theory and disability, and concepts of NARIC Accession Number: J53625 Project Number: H133N060014 voicelessness and personhood. Future directions for ABSTRACT: Article presents results from a pilot BCI research are recommended based on the emergent clinical evaluation of a BrainGate (Cyberkinetics, focus group themes. Page 9 Inc.), a neuromotor prosthesis (NMP) in a man with June 9-14, 1995, Vancouver, BC, Canada), ed. A tetraplegia. An NMP is a type of brain-computer Langton, Arlington, VA: RESNA Press, 684-686. interface designed to restore lost motor function NARIC Accession Number: O12728 in paralyzed humans by routing movement-related ABSTRACT: Paper presenting a case description of signals from the brain to external effectors. NMPs the use of a brain-computer interface (BCI) by a man require that intention-driven neuronal activity be with amyotrophic lateral sclerosis (ALS). The man converted into a control signal that enables useful learned to use EEG activity over the sensorimotor tasks. Neuronal ensemble activity recorded through a cortex to control cursor movement on a video screen. 96-electrode array implanted in the subject’s primary His development of control was facilitated by off-line motor cortex demonstrated that intended hand motion topographic and frequency analysis of EEG data re- modulates cortical spiking patterns three years after corded during training. This analysis led to improve- spinal cord injury. Decoders were created to produce a ments in the online algorithm that converted EEG neural cursor with which the subject opened simulated activity into cursor movement, and enabled the user e-mail and operated devices such as a television. The to move the cursor more rapidly and accurately. The subject also used neural control to open and close a authors conclude that BCI technology might be an prosthetic hand, and perform basic tasks with a multi- alternative communication option for those with ALS. jointed robotic arm. This paper was presented at the 1995 annual conference of the Rehabilitation Engineering and Assistive 2001 Technology Society of North America (RESNA).

Birbaumer, N.P., Hinterberger, T., Kaiser, J., Ko- tchoubey, B., Kubler, A., & Neumann, N. (2001). Brain-computer communication: Self-regulation of slow cortical potentials for verbal communica- Full-text copies of these documents may be avail- tion. Archives of Physical Medicine and Rehabilita- able through NARIC’s document delivery service. tion, 82(11), 1533-1539. NARIC Accession Number: J43250 To order any of the documents listed above, note the ABSTRACT: Article describes the validation of a accession number and call an information special- training procedure designed to enable verbal com- ist at 800/346-2742. munication through self-regulation of slow cortical potentials. Two male patients with late-stage There is a charge of 5 cents for copying and ship- amyotrophic lateral sclerosis (ALS) learned voluntary ping with a $5 minimum on all orders. control of their slow cortical potentials using a brain- computer interface wherein electric brain activity is used to control a cursor on a computer screen. Within 3 to 8 weeks, both had learned to use this skill to select letters or words and communicate them. The protocol may be a model for training other brain-computer interface techniques.

1995

McCane, L., McFarland, D.J., Vaughan, T.M., & Wolpaw, J.R. (1995). Development of direct brain-computer communication in a person with Amyotrophic Lateral Sclerosis. RECREAbility - Recreation and Ability: Explore the Possibilities! (Proceedings of the RESNA ‘95 Annual Conference, Page 10 Documents from the Center square (RLS) algorithm that describes the current for International state of the EEG in real-time. First results of the new Rehabilitation Research low-cost BCI show that the accuracy of differentiat- Information and Exchange ing imagination of left and right hand movement is (CIRRIE-3) search at cirrie. around 95 percent. buffalo.edu are listed below: 1996 2002 Homann, N., Kalcher, J., Neuper, C., & Rothl, W., et Karmali, F., Kostov, A., & Polak, M. (2002). Envi- al. (1996). Graz brain-computer interface (BCI): ronmental control by a brain-computer interface. A pilot study on patients with motor deficits. In- Engineering in Medicine and Biology Society, terdisciplinary Aspects on Computers Helping People 2000. Proceedings of the 22nd Annual International with Special Needs, 1, 155-60. Conference of the IEEE, 4, 3001-3. ABSTRACT: The paper describes the EEG-based ABSTRACT: In this paper a system is described Brain Computer Interface (BCI) in which the EEG which uses a brain-computer interface (BCI) and drives a communication channel between the brain thus enables the persons with amyotrophic lateral and an electrical appliance. A pilot study was con- sclerosis (ALS) to control home appliances, and to ducted to explore the use of this method with an use computer with speech programs to communicate equipment used to test persons with motor difficul- through articulation of pre-selected words. The Ac- ties. The results demonstrated that the imagination tiveHome from X10, which controls various home of movement results creates similar EEG patterns as appliances through a computer, is used. The ease-of- does the preparation of a real movement. (CIRRIE use, ease-of-learning, rate-of-error, and amount of Abstract). time required for task completion, were evaluated. (CIRRIE Abstract).

1999 Documents from the Education Guger, C., Pfurtscheller, G., Schlogl, A., & Wal- Resource Information Center terspacher, D. (1999). Design of an EEG-based (ERIC) search at www.eric.ed.gov brain-computer interface (BCI) from standard are listed below: components running in real-time under Windows. Biomedizinische Technik, 44(1-2), 12-6. 2013 ABSTRACT: An EEG-based brain-computer interface (BCI) is a direct connection between the Chau, T., Guerguerian, A.M., Kushki, A., & Moghimi, human brain and the computer. Such a communica- S. (2013). A review of EEG-based brain-computer tion system is needed by patients with severe motor interfaces as access pathways for individuals with impairments (e.g. late stage of Amyotrophic Lateral severe disabilities. Assistive Technology, 25(2), Sclerosis) and has to operate in real-time. This paper 99-110. describes the selection of the appropriate compo- ERIC Number: EJ1005025 nents to construct such a BCI and focuses also on Electroencephalography (EEG) is a non-invasive the selection of a suitable programming language method for measuring brain activity and is a strong and operating system. The multichannel system runs candidate for brain-computer interface (BCI) devel- under Windows 95, equipped with a real-time Kernel opment. While BCIs can be used as a means of com- expansion to obtain reasonable real-time operations munication for individuals with severe disabilities, on a standard PC. Matlab controls the data acquisition the majority of existing studies have reported BCI and the presentation of the experimental paradigm, evaluations by able-bodied individuals. Considering while Simulink is used to calculate the recursive least the many differences in body functions and usage Page 11 scenarios between individuals with disabilities and Documents from the Cochrane able-bodied individuals, involvement of the target Database of Systematic Reviews search population in BCI evaluation is necessary. In this at www.thecochranelibrary.org are review, 39 studies reporting EEG-oriented BCI as- listed below: sessment by individuals with disabilities were identified in the past decade. With respect to participant 2014 populations, a need for assessing BCI performance Haider, S., Heckman, S., Hill, N.J., McCane, L.M., for the pediatric population with severe disabilities Ricci, E., Vaughan, T.M., & Wolpaw, J.R. (2014). A was identified as an important future direction. Ac- practical, intuitive brain-computer interface for quiring a reliable communication pathway during communicating ‘yes’ or ‘no’ by listening. Journal early stages of development is crucial in avoiding of Neural Engineering, 11(3), 035003. learned helplessness in pediatric-onset disabilities. ID: CN-01053775 With respect to evaluation, augmenting traditional ABSTRACT: OBJECTIVE: Previous work has measures of system performance with those relating shown that it is possible to build an EEG-based binary to contextual factors was recommended for realizing brain-computer interface system (BCI) driven purely user-centered designs appropriate for integration in by shifts of attention to auditory stimuli. However, real-life. Considering indicators of user state and previous studies used abrupt, abstract stimuli that are developing more effective training paradigms are often perceived as harsh and unpleasant, and whose recommended for future studies of BCI involving lack of inherent meaning may make the interface individuals with disabilities. unintuitive and difficult for beginners. We aimed to -es tablish whether we could transition to a system based 1993 on more natural, intuitive stimuli (spoken words ‘yes’ Rechlin, T. (1993). Communication system in cases and ‘no’) without loss of performance, and whether of “locked-in” syndrome. International Journal of the system could be used by people in the locked-in Rehabilitation Research, 16(4), 340-42. state. APPROACH: We performed a counterbalanced, ERIC Number: EJ477675 interleaved within-subject comparison between an This paper describes a computer-assisted communica- auditory streaming BCI that used beep stimuli, and tion system successfully used in three cases of “locked one that used word stimuli. Fourteen healthy volun- in syndrome” (in which the only reliable motor func- teers performed two sessions each, on separate days. tions are eye movements). The paper concludes that We also collected preliminary data from two subjects patients who survive locked in syndrome for several with advanced amyotrophic lateral sclerosis (ALS), weeks should be given the chance to communicate who used the word-based system to answer a set of via a suitable system. simple yes-no questions. MAIN RESULTS: The N1, N2 and P3 event-related potentials elicited by words varied more between subjects than those elicited by beeps. However, the difference between responses to attended and unattended stimuli was more consistent with words than beeps. Healthy subjects’ performance with word stimuli (mean 77 percent ± 3.3 s.e.) was slightly but not significantly better than their performance with beep stimuli (mean 73 percent ± 2.8 s.e.). The two subjects with ALS used the word-based BCI to answer questions with a level of accuracy similar to that of the healthy subjects. SIGNIFICANCE: Since performance using word stimuli was at least as good as performance using beeps, we recommend that auditory streaming BCI systems be built with Page 12 word stimuli to make the system more pleasant and Documents from the IEEE Xplore intuitive. Our preliminary data show that word-based Digital Library search at ieeexplore. streaming BCI is a promising tool for communication ieee.org are listed below: by people who are locked in. 2014 2004 Birch, G.E., Bohringer, R., Borisoff, J.F., & Mason, Ibrahim, B.S.K.K., & Sherwani, F.. (2014). Brain S.G. (2004). Real-time control of a video game computer interface based functional electrical with a direct brain--computer interface. Journal stimulation: An outline. Functional Electrical of Clinical Neurophysiology, 21(6), 404-8. Stimulation Society Annual Conference (IFESS), ID: CN-00502182 2014 IEEE 19th International, 1-5, 17-19 Sept. 2014. ABSTRACT: Mason and Birch have developed a di- doi: 10.1109/IFESS.2014.7036766. rect brain-computer interface for intermittent control ABSTRACT: People affected by spinal cord injury of devices such as environmental control systems and (SCI) are usually unable to move their lower limbs neuroprotheses. This EEG-based brain switch, named due to inactive control of the muscles from the brain. the LF-ASD, has been used in several off-line studies, This lack of movement may lead to further moral and but little is known about its usability with real-world physical complexities such as cardiovascular diseases, devices and computer applications. In this study, bone demineralization and bedsores. Physiotherapy able-bodied individuals and people with high-level based exercise and training are conventionally ad- spinal injury used the LF-ASD brain switch to con- vised to such plegic patients, which, hereby, has not trol a video game in real time. Both subject groups been shown to be have ample recovery efficiency. demonstrated switch activations varying from 30 Alternatively, Functional electrical stimulation (FES) percent to 78 percent and false-positive rates in the is a relatively newer technique which uses electrical signals to energize the neurons and excite the tissues range of 0.5 percent to 2.2 percent over three 1-hour in the muscles while producing the corresponding test sessions. These levels correspond to switch clas- contraction in them. FES alone however requires sification accuracies greater than 94 percent for all specific electrical devices to generate and supply subjects. The results suggest that subjects with spinal certain electrical signals similar to that of generated cord injuries can operate the brain switch to the same by human brain. This needs some additional devices ability as able-bodied subjects in a real-time control to be used as the control system for FES to identify environment. These results support the findings of and issue the commands as required from time to previous studies. time. A brain-computer interface (BCI) is a direct communication pathway between the brain and an external device. It uses electrodes, placed on the scalp, to collect signals from the brain structure. A combination of BCI and FES can be a vital solution to cater this issue, as the paralyzed patient can use his own brain electroencephalogram (EEG) as a control system to perform the required movements. This paper discusses their advantages, short comings and latest research advances in this field. Firstly, the significance of FES devices is being introduced and the different technological techniques reported in literature are discussed. Secondly, human brain is introduced as a control system to be employed within BCI systems to generate the required EEG signal activity. Finally, an incorporation of both FES and BCI is suggested to overcome the presiding issues regarding efficient control of the muscles. Page 13 2013 what is called “locked-in syndrome.” Consequently, a Brain Computer Interface (BCI) can be used as Huang, T.W., Sun, K.T., Tai, Y.H., & Tian, Y.J. (2013). an alternative communication channel. This project Brainwave technology gives Internet access to the belongs to a Brain Computer Interface research. physically disabled. Intelligent Systems (GCIS), More precisely, it focuses on the development of 2013 Fourth Global Congress, 331-335, 3-4 Dec. noninvasive platform of electroencephalographic 2013. signals in terms of acquisition, pre-processing, feature doi: 10.1109/GCIS.2013.59. extraction for providing an alternative communica- ABSTRACT: It is impossible for severely disabled tion or control channel for patient with severe motor people to browse or learn through the Internet due to disabilities. the mere lack of independent control of the mouse. This paper proposes a brain computer interface (BCI) 2012 to aid severely disabled individuals, such as people disabled by amyotrophic lateral sclerosis, in brows- Lijun, J., Ong, G.P., Tham, E., & Yeo, M. (2012). ing or learning on the Internet. By analyzing specific iPhone-based portable brain control wheelchair. components of event-related potentials, cursor control Industrial Electronics and Applications (ICIEA), 2012 can be achieved. The cursor can be controlled by 7th IEEE Conference, 1592-1594, 18-20 July 2012. brain waves in a user-friendly manner to move or doi: 10.1109/ICIEA.2012.6360978. click on the web page. The major contributions of ABSTRACT: A novel system which uses brain this research include: (1) designing a BCI for disabled computer interface (BCI) technology and an iPhone people, (2) embedding the BCI’s cursor controller into to control wheelchair is developed. This system is a web page, (3) conducting clinical experiments, (4) developed with an aim to assist people suffering from analyzing the experimental results, system accuracy, spinal cord injuries, and amyotrophic lateral sclerosis, and effectiveness, and (5) evaluating the system’s due to the inability to move limbs and body caused practicability and giving suggestions for future work. by death of nerve cells. The proposed system uses an There are two innovative technologies proposed iPhone to process the brain signals (EEG), specifi- in this research: (1) a specific component of ERPs cally the attention signals, and direct corresponding located at O1, the N2P3 (the difference between the commands to an assistive device, such as wheelchair, peak and trough of N200 and P300), was used to dif- wirelessly. Comparing to other BCI system comprises ferentiate targeted from non-targeted (non-selected) of a bulky laptop and multi-electrodes, the proposed signals, (2) instead of a fixed position style BCI, a system is portable, cost-effective and ease of use. movable watermark style BCI which follows the cur- Potentially, not only people who are suffering from sor was designed. The novel technique supported by a spinal cord injuries can benefit from this system, user-friendly interface helps the disabled have contact it extended to assisting elderly or people suffering with the outside world resulting in positive emotions. from limb muscle disorder. The system structure and experimental results will be illustrated in this paper. Rihana, S., & Saker, M. (2013). Platform for EEG No similar system is reported yet. signal processing for motor imagery: Application brain computer interface. Advances in Biomedi- 2011 cal Engineering (ICABME), 2013 2nd International Conference, 30-33, 11-13 Sept. 2013. Ali, M., Imran, M., Rizwan, M., & Ullah, K. (2011). doi: 10.1109/ICABME.2013.6648839. Low-cost single-channel EEG based communica- ABSTRACT: Over two million people are affected tion system for people with locked-in (sic) syn- by neural diseases such as multiple sclerosis, amyo- drome. Multitopic Conference (INMIC), 2011 IEEE trophic lateral sclerosis, spinal cord injury, cerebral 14th International, 120-125, 22-24 Dec. 2011. palsy, and other diseases impairing the neural path- doi: 10.1109/INMIC.2011.6151455. ways that control muscles. Indeed, these diseases ABSTRACT: As the use of biomedical signals is cause severe paralysis and the persons suffer from incredibly increasing in both clinical and nonclinical Page 14 applications. They have a great deal in the develop- 2010 ment of devices that can be controlled by information inferred from thoughts. One of the current hot topics Adom, A.H., Hema, C.R., Paulraj, M.P., & Purusho- for research is the Brain Computer Interface (BCI) on thaman, D. (2010). Brain machine interface for basis of EEG signals. BCI is a technology that makes physically retarded people using color visual tasks. humans to control computer or other devices on basis Signal Processing and Its Applications (CSPA), 2010 of information inferred from thoughts. BCIs have 6th International Colloquium, 1-4, 21-23 May 2010. given new hopes to people who suffer from locked-in doi: 10.1109/CSPA.2010.554533. syndrome and motor disabilities by providing alterna- ABSTRACT: A Brain Machine Interface is a com- tive means of communication channels. The existing munication system which connects the human brain BCIs are Multi-channel, thus very expensive in terms activity to an external device bypassing the peripheral of cost and processing speed, which make them dif- nervous system and muscular system. It provides ficult for domestic use. The aim of this research paper a communication channel for the people who are is to introduce a very cheap, simple and a robust single suffering with neuromuscular disorders such as channel BCI that could prevail in the market. We pro- amyotrophic lateral sclerosis, brain stem stroke, quad- pose a very low-cost EEG-based BCI that is designed riplegics and spinal cord injury. In this paper, a simple to help severely disabled people communicate with BMI system based on EEG signal emanated while others by means of text and SMS. To make it simple visualizing of different colors has been proposed. The and affordable, the number of channels is limited to proposed BMI uses the color visual tasks and aims one and signal is acquired through homemade silver to provide a communication through brain activated electrodes and then fed to the computer through the control signal for a system from which the required soundcard for further processing and features extrac- task operation can be performed to accomplish the tions. The experimental results show that the proposed needs of the physically retarded community. The system is capable enough to provide a very low cost, ability of an individual to control his EEG through yet reliable, communication means and a suitable BCI the color visualization enables him to control devices. for domestic use. Its average accuracy is 87 percent. The EEG signal is recorded from 10 voluntary healthy The potential uses for the technology are almost subjects using the noninvasive scalp electrodes placed limitless. Instead of communication system, disabled over the frontal, parietal, motor cortex, temporal users could have robotic wheelchair, allowing them to and occipital areas. The obtained EEG signals were segmented and then processed using an elliptic filter. move and directly interact with the environments thus Using spectral analysis, the alpha, beta and gamma it can be used for clinical and nonclinical purposes. band frequency spectrum features are obtained for each EEG signals. The extracted features are then Perkusich, A., Perkusich, M.B., & Rached, T.S. associated to different control signals and a neural (2011). ThinkContacts: Use your mind to dial your network model using back propagation algorithm has phone. Consumer Electronics (ICCE), 2011 IEEE been developed. The proposed method can be used to International Conference, 105-106, 9-12 Jan. 2011. translate the color visualization signals into control doi: 10.1109/ICCE.2011.5722486. signals and used to control the movement of a mobile ABSTRACT: Integration between Brain Computer robot. The performance of the proposed algorithm has Interface and mobile devices allows the development an average classification accuracy of 95.2 percent. of mobile applications for people with severe neuromuscular diseases. These applications can increase Almeida, H., de Almeida Holanda, M.M., de S their quality of life. This paper presents a mobile ap- Santos, D.F., Perkusich, A., & Rached, T.S. (2010). plication that enables disabled people to make a phone BCI-aware pervasive multimedia for motor dis- call to a desired contact on their phone’s address book abled people. Information Society (i-Society), 2010 solely using their brain activity. International Conference, 86-91, 28-30 June 2010. ABSTRACT: Assistive technologies have a major impact in the life of people with severe neuromuscu- Page 15 lar diseases in order to provide or extend functional trial EEG. Results suggest that this method may be abilities of people with special needs. Such technolo- used to enhance control of HMI’s for individuals with gies promote independent life, social inclusion and severe mobility limitations. improve quality of life of disabled people. Brain- Computer Interface (BCI) is a recent technology that Hongtao, W., Peng, C., Ting, L., & Yihong, H. (2010). allows the direct communication between the brain Satellite television remote control system based on and the environment. The development of systems brain-computer interface. Intelligent Computation that integrates BCI, smart devices and pervasive Technology and Automation (ICICTA), 2010 Inter- services enables motor disabled people to control national Conference, 3, 264-267, 11-12 May 2010. devices anytime and anywhere. This paper presents doi: 10.1109/ICICTA.2010.819. an architecture for the development of BCI-aware ABSTRACT: This paper presents the satellite televi- pervasive systems to control multimedia devices. The sion remote control system based on brain-computer user needs an electroencephalography device and a interface. The Brain Controlled Satellite Television smart device to interact with any multimedia device. Remote System (BCSTRS) is a real time system that This paper brings the implementation of a BCI-aware can help the patients suffering from Amyotrophic pervasive system that enables users to control the Lateral Sclerosis (ALS) to select TV channels or ad- XBMC media center to play music or videos. just volume using their brain waves. In this paper we propose an algorithm including data acquisition and Arshak, K., Burke, E., Condron, J., & Kaneswaran, process, feature extraction, pattern recognition and K. (2010). Towards a brain controlled assistive SVM classifier. Experiments demonstrated that the technology for powered mobility. Engineer- BCSTRS is able to achieve an averaged information ing in Medicine and Biology Society (EMBC), transfer rate of approximately 18 b/min and 5 healthy 2010 Annual International Conference of the subjects can control the BCSTRS efficiently with an IEEE, 4176-4180, Aug. 31 2010-Sept. 4 2010. average accuracy of 90 percent. doi: 10.1109/IEMBS.2010.5627385 ABSTRACT: For individuals with mobility limita- In-Ho, L., Jong-Bae, K., Kwang-Ok, A., & Won- tions, powered wheelchair systems provide improved Kyoung, S. (2010). Development of an emer- functionality, increased access to healthcare, educa- gency call system using a brain computer interface tion and social activities. Input devices such as joy- (BCI). Biomedical Robotics and Biomechatronics rd stick and switches can provide the necessary input re- (BioRob), 2010 3 IEEE RAS and EMBS Inter- quired for efficient control of the powered wheelchair. national Conference, 918-923, 26-29 Sept. 2010. For persons with limited dexterity, or fine control of doi: 10.1109/BIOROB.2010.5626331. the fingers, access to mechanical hardware such as ABSTRACT: A brain computer interface (BCI) can buttons and joysticks can be quite difficult and some- be used by persons with severe neuromuscular dis- times painful. For individuals with conditions such as orders to control external devices such as computers Traumatic Brain Injury, Multiple Sclerosis or Amyo- or neuro-prosthesis. To improve accuracy, however, trophic lateral sclerosis voluntary control of limb most of BCIs need multi-channels, wet sensors, big movement maybe substantially limited or completely and complex systems. It makes most use of BCIs absent. Brain Computer Interfaces are emerging as a restricted to laboratory and medical area. In this possible method to replace the brains normal output paper, to solve the limitation, mindset (neurosky pathways of peripheral nerves and muscles, allowing headset) that is a commercially available BCI is individuals with paralysis a method of communication adopted and the emergency call system for severely and computer control. This study involves the analysis disabled persons using attention/meditation signal is of non-invasive electroencephalograms (EEG) arising developed. In addition, to operate on a wide range from the use of a newly developed Human Machine of disabilities under a wide range of environmental Interface for powered wheelchair control. Using a conditions, algorithms of determining trigger method delayed response task, binary classification of left and threshold level are proposed. The usefulness of and right movement intentions were classified with the system and the proposed algorithms is verified by a best classification rate of 81.63 percent from single experiment result. Page 16 Rao, R.P.N., & Scherer, R. (2010). Brain-Com- use a remote robot controlled by a highly paralyzed puter Interfacing [In the Spotlight]. Signal patient via a BCI through a friendly graphic user. Processing Magazine, IEEE , 27(4), 152-150. Some preliminary experiments are presented in this doi: 10.1109/MSP.2010.936774. paper about one of the possible application: a robotic ABSTRACT: Recently, CNN reported on the future museum guide (PeopleBot and Pioneer3 robot), that of brain-computer interfaces (BCIs). BCIs are devices can transmit remote visual perceptions to the patient. that process a user’s brain signals to allow direct communication and interaction with the environ- 2008 ment. BCIs bypass the normal neuromuscular output pathways and rely on digital signal processing and Inoue, T., Kamata, M., Madarame, T., Shino, M., & machine learning to translate brain signals to action Tanaka, H. (2008). The development of a brain (Figure 1). Historically, BCIs were developed with computer interface device for amyotrophic lateral biomedical applications in mind, such as restoring sclerosis patients. Systems, Man, and Cybernetics, communication in completely paralyzed individuals 2008. SMC 2008. IEEE International Conference, and replacing lost motor function. More recent ap- 2401-2406, 12-15 Oct. 2008. plications have targeted nondisabled individuals by doi: 10.1109/ICSMC.2008.4811654 exploring the use of BCIs as a novel input device for ABSTRACT: The objective of this research was entertainment and gaming. The task of the BCI is to to develop a brain computer interface (BCI) com- identify and predict behaviorally induced changes munication device for amyotrophic lateral sclerosis or “cognitive states” in a user’s brain signals. Brain (ALS) patients. The device was designed to meet the signals are recorded either noninvasively from elec- needs of ALS patients, and to be used at a clinical trodes placed on the scalp [electroencephalogram] or level. Initial tests were performed by ALS patients, invasively from electrodes placed on the surface of or and the result was accounted for in the experimental inside the brain. BCIs based on these recording tech- production of the communication device. Lastly, the niques have allowed healthy and disabled individuals device was evaluated by able-bodied examinees and to control a variety of devices. In this article, we will ALS patients. For able-bodied examinees, the device describe different challenges and proposed solutions scored a high rate of correct sessions. When an ALS for noninvasive brain-computer interfacing. patient was the user, the correction rate was not as well, but it would have scored highly if a correct 2009 parameter was chosen.

Anzalone, S.M., Blanda, C., Buttita, E., Chella, 2007 A., Cinquegrani, F., Menegatti, E., Pagello, E., Piccione, F., Prifitis, K., Sorbello, R., Tonin, L., & Ang, V.M.H., Burdet, E., Chee, L.T., Guan, C., Tranchina, E. (2009). A BCI teleoperated mu- Laugier, C., Rebsamen, B., Zeng, Q., & Zhang, H. seum robotic guide. Complex, Intelligent, and (2007). Controlling a wheelchair indoors using Software Intensive Systems, 2009. CISIS ‘09. Inter- thought. Intelligent Systems, IEEE, 22(2), 18-24. national Conference, 783-788, 16-19 March 2009. doi: 10.1109/MIS.2007.26 doi: 10.1109/CISIS.2009.154. ABSTRACT: Amyotrophic lateral sclerosis, or ALS, ABSTRACT: Brain computer interface is a system is a degenerative disease of the motor neurons that that offers also a support to the patients with neuro- eventually leads to complete paralysis. We are devel- muscular diseases as amyotrophic lateral sclerosis. In oping a wheelchair system that can help ALS patients, this paper are presented some works with the aim to and others who can’t use physical interfaces such as integrate brain computer interfaces and mobile robots. joysticks or gaze tracking, regain some autonomy. The two aim of this project are: (1) to test an improved The system must be usable in hospitals and homes BCI experience through the help of a physical robot, with minimal infrastructure modification. It must be so that brain signals are stronger stimulate, and (2) to safe and relatively low cost and must provide optimal Page 17 interaction between the user and the wheelchair within a computer using EEG. A Brain Computer Interface the constraints of the brain-computer interface. To is a communication system in which messages or this end, we have built the first working prototype of commands that an individual sends to the external a brain-controlled wheelchair that can navigate inside world do not pass through the brain’s normal output a typical office or hospital environment. This article pathways of peripheral nerves and muscles. This describes the BCW, our control strategy, and the sys- type of interface would increase an individual’s in- tem’s performance in a typical building environment. dependence, leading to an improved quality of life This brain-controlled wheelchair prototype uses a and reduced social costs. A system is created to allow P300 EEG signal and a motion guidance strategy to individuals, with the help of BCI, to control wide va- navigate in a building safely and efficiently without riety of home appliances via a computer, and allows complex sensors or sensor processing. tasks such as light switching and turning appliances on and off. Ortiz, Jr, S. (2007). Brain-computer interfaces: Where human and machine meet. Computer, 40(1), 2006 pp.17-21. doi: 10.1109/MC.2007.11. Krusienski, D.J., McFarland, D.J., Sarnacki, W.A., ABSTRACT: For a long time, researchers have been Schalk, G., Sellers, E.W., Vaughan, T.M., & Wolpaw, working on a marriage of human and machine that J.R. (2006). The Wadsworth BCI Research and sounds like something out of science fiction: a brain- Development Program: At home with BCI. Neu- computer interface. BCIs read electrical signals or ral Systems and Rehabilitation Engineering, IEEE other manifestations of brain activity and translate Transactions, 14(2), 229-233. them into a digital form that computers can under- doi: 10.1109/TNSRE.2006.875577. stand, process, and convert into actions of some kind, ABSTRACT: The ultimate goal of brain-computer such as moving a cursor or turning on a TV. Several interface (BCI) technology is to provide communi- academic and corporate researchers are now working cation and control capacities to people with severe to commercialize the technology, while other projects motor disabilities. BCI research at the Wadsworth are taking innovative approaches to BCIs that could Center focuses primarily on noninvasive, electroen- create interesting products or services in the not-too- cephalography (EEG)-based BCI methods. We have distant future. The technology holds great promise for shown that people, including those with severe motor people who can’t use their arms or hands normally disabilities, can learn to use sensorimotor rhythms because they have had spinal cord injuries or suffer (SMRs) to move a cursor rapidly and accurately from conditions such as amyotrophic lateral sclerosis in one or two dimensions. We have also improved or cerebral palsy. BCI could help them control com- P300-based BCI operation. We are now translating puters, wheelchairs, televisions, or other devices with this laboratory-proven BCI technology into a system brain activity. that can be used by severely disabled people in their homes with minimal ongoing technical oversight. Pahuja, R. Pahuja, S., & Singla, R. (2007). Envi- To accomplish this, we have improved our general- ronment control using BCI. Bioinformatics and purpose BCI software (BCI2000), online adaptation Biomedical Engineering, 2007. ICBBE 2007. The 1st and feature translation for SMR-based BCI operation, International Conference, 1293-1295, 6-8 July 2007. the accuracy and bandwidth of P300-based BCI op- doi: 10.1109/ICBBE.2007.333. eration; reduced the complexity of system hardware ABSTRACT: Brain-Computer Interfaces (BCI) are and software and begun to evaluate home system developed to help locked-in patients, who lose con- use in appropriate users. These developments have trol of their bodies and are unable to perform simple resulted in prototype systems for everyday use in tasks such as speech, locomotion, and can’t even ef- people’s homes. fectively interact, with their environment. BCI shows promise in allowing these individuals to interact with Page 18 2003 Engineering in Medicine and Biology Society, 2000. Proceedings of the 22nd Annual International Con- Kubler, A., & Neumann, N. (2003). Training ference of the IEEE, 4, 2990-2992. locked-in patients: A challenge for the use of doi: 10.1109/IEMBS.2000.901508. brain-computer interfaces. Neural Systems and ABSTRACT: Locked-in patients, such as those Rehabilitation Engineering, IEEE Transactions, with amyotrophic lateral sclerosis, lose control of 11(2), 169-172. their bodies, leaving them unable to perform simple doi: 10.1109/TNSRE.2003.814431. tasks such as speech, locomotion, and the ability to Abstract: Training severely paralyzed patients to use effectively interact with their environment. Brain- a brain-computer interface (BCI) for communication Computer Interfaces (BCI) show promise in allowing poses a number of issues and problems. Over the past these individuals to interact with a computer using six years, we have trained 11 patients to self-regulate EEG. A system was created to allow individuals, their slow cortical brain potentials and to use this skill via a BCI, to control home appliances and to have to move a cursor on a computer screen. This paper the computer articulate pre-selected words. It uses describes our experiences with this patient group ActiveHome from X10, which can control of a wide including the problems of accepting and rejecting variety of home appliances via a computer, and allows patients, communicating and interacting with patients, tasks such as light dimming, TV channel changing, how training may be affected by social, familial, and and turning appliances on and off. The system was institutional circumstances, and the importance of evaluated with a set of tasks to measure its ease-of- motivation and available reinforcers. use, ease-of-learning, rate-of-error, and amount of time the subject required to complete the task. Moore, M.M. (2003). Real-world applications for brain-computer interface technol- 1998 ogy. Neural Systems and Rehabilitation Engi- neering, IEEE Transactions, 11(2), 162-165. Kostov, A. (1998). Machine learning meth- doi: 10.1109/TNSRE.2003.814433. ods in assistive technologies. Systems, Man, ABSTRACT: The mission of the Georgia State Uni- and Cybernetics, 1998. 1998 IEEE International versity Brain Lab is to create and adapt methods of Conference, 4, 3729-3734. 11-14 Oct 1998. human-computer interaction that will allow brain- doi: 10.1109/ICSMC.1998.726667. computer interface (BCI) technologies to effectively ABSTRACT: Assistive devices are essential in en- control real-world applications. Most of the existing hancing the quality of life for individuals who have BCI applications were designed largely for training severe disabilities, such as quadriplegia and amyo- and demonstration purposes. Our goal is to research trophic lateral sclerosis, or who have had massive ways of transitioning BCI control skills learned in brainstem strokes. However, the effectiveness of these training to real-world scenarios. Our research ex- systems is dependent on preserved residual move- plores some of the problems and challenges of comments or speech. In the absence of means to repair bining BCI outputs with human-computer interface the damaged nervous system, three options exist for paradigms in order to achieve optimal interaction. We restoring function: (1) augmenting the capabilities of utilize a variety of application domains to compare remaining pathways, (2) detouring around points of and validate BCI interactions, including communica- damage, or (3) providing the brain with new channels tion, environmental control, neural prosthetics, and for communication and control. The paper reviews creative expression. The goal of this research is to im- the use of machine learning methods for development prove quality of life for those with severe disabilities. of assistive technology. Three projects are described, representing the three options listed above. In each 2000 of them machine learning methods are employed to help with pattern recognition and classification. The Karmali, F., Kostov, A., & Polak, M. (2000). Envi- three projects are: automatic speech recognition of ronmental control by a brain-computer interface. dysarthric speech, control strategies for FES-assisted Page 19 locomotion (functional electrical stimulation), and an Documents from the National EEG-based computer access. Although these three Library of Medicine PubMed projects may look very different from each other, search at www.pubmed.com are the structure of their experimental set-ups, and their listed below: potential for application in assistive devices are very similar. All experimental set-ups consist of sensory 2015 signal acquisition, signal processing for feature extraction, and data processing by machine learning Birbaumer, N., Chaudhary, U., & Curado, M.R. techniques for pattern recognition and classification. (2015). Brain-machine interface (BMI) in paraly- In addition, all three projects deal with digital signal sis. Annals of Physical and Rehabilitation Medicine, processing and machine learning method applications 58(1), 9-13. in development of man-machine interfaces. PMID: 25623294 ABSTRACT: INTRODUCTION: Brain-machine interfaces (BMIs) use brain activity to control external devices, facilitating paralyzed patients to interact with the environment. In this review, we focus on the current advances of non-invasive BMIs for communication in patients with amyotrophic lateral sclerosis (ALS) and for restoration of motor impairment after severe stroke. BMI FOR ALS PATIENTS: BMI represents a promising strategy to establish communication with paralyzed ALS patients as it does not need muscle engagement for its use. Distinct techniques have been explored to assess brain neurophysiology to control BMI for patients’ communication, especially electroencephalography (EEG) and more recently near-infrared spectroscopy (NIRS). Previ- ous studies demonstrated successful communication with ALS patients using EEG-BMI when patients still showed residual eye control, but patients with complete paralysis were unable to communicate with this system. We recently introduced functional NIRS (fNIRS)-BMI for communication in ALS patients in the complete locked-in syndrome (i.e., when ALS patients are unable to engage any muscle), opening new doors for communication in ALS patients after complete paralysis. BMI FOR STROKE MOTOR RECOVERY: In addition to assisted communication, BMI is also being extensively studied for motor recovery after stroke. BMI for stroke motor recovery includes intensive BMI training linking brain activity related to patient’s intention to move the paretic limb with the contingent sensory feedback of the paretic limb movement guided by assistive devices. BMI studies in this area are mainly focused on EEG- or magnetoencephalography (MEG)-BMI systems due to their high temporal resolution, which facilitates online contingency between intention to move and sen- Page 20 sory feedback of the intended movement. EEG-BMI PMID: 25959328 training was recently demonstrated in a controlled ABSTRACT: By focus group methodology, we study to significantly improve motor performance examined the opinions and requirements of persons in stroke patients with severe paresis. Neural basis with ALS, their caregivers, and health care assistants for BMI-induced restoration of motor function and with regard to developing a brain-computer interface perspectives for future BMI research for stroke motor (BCI) system that fulfills the user’s needs. Four recovery are discussed. overarching topics emerged from this analysis: (1) lack of information on BCI and its everyday applica- Blandin, V., Bruno, M.A., Demertzi, A., Gosseries, tions, (2) importance of a customizable system that O., Heine, L., Laureys, S., Lugo, Z.R., Pellas, F., supports individuals throughout the various stages of & Thonnard, M. (2015). Beyond the gaze: Com- the disease, (3) relationship between affectivity and municating in chronic locked-in syndrome. Brain technology use, and (4) importance of individuals Injury, 29(9), 1056-61. retaining a sense of agency. These findings should be PMID: 26182228 considered when developing new assistive technol- ABSTRACT: OBJECTIVE: Locked-in syndrome ogy. Moreover, the BCI community should acknowl- (LIS) usually follows a brainstem stroke and is edge the need to bridge experimental results and its characterized by paralysis of all voluntary muscles everyday application. (except eyes’ movements or blinking) and lack of speech with preserved consciousness. Several tools Geronimo, A., Schiff, S.J., Simmons, Z., & Stephens, have been developed to promote communication with H.E. (2015). Acceptance of brain-computer inter- these patients. The aim of the study was to evaluate the current status regarding communication in a cohort faces in amyotrophic lateral sclerosis. Amyotrophic of LIS patients. DESIGN: A survey was conducted in Lateral Sclerosis & Frontotemporal Degeneration, collaboration with the French Association of Locked- 16(3-4), 258-64. in syndrome (ALIS). SUBJECTS AND METHODS: PMID: 25372874 Two hundred and four patients, members of ALIS, ABSTRACT: Brain-computer interfaces (BCI) have were invited to fill in a questionnaire on communica- the potential to permit patients with amyotrophic tion issues and clinical evolution (recovery of verbal lateral sclerosis (ALS) to communicate even when language and movements, presence of visual and/or locked in. Although, as many as half of patients with auditory deficits). RESULTS: Eighty-eight responses ALS develop cognitive or behavioral dysfunction, the were processed. All respondents (35 percent female, impact of these factors on acceptance of and ability to mean age = 52 ± 12 years, mean time in LIS = 10 ± 6 use a BCI has not been studied. We surveyed patients years) reported using a yes/no communication code with ALS and their caregivers about BCIs used as using mainly eyes’ movements and 62 percent used assistive communication tools. The survey focused assisting technology; 49 percent could communicate on the features of a BCI system, the desired end-use through verbal language and 73 percent have recov- functions, and requirements. Functional, cognitive, ered some functional movements within the years. and behavioral data were collected from patients and CONCLUSION: The results highlight the possibility analyzed for their influence over decisions about BCI to recover non-eye dependent communication, speech device use. Results showed that behavioral impair- production and some functional movement in the ment was associated with decreased receptivity to the majority of chronic LIS patients. use of BCI technology. In addition, the operation of a BCI system during a pilot study altered patients’ Cincotti, F., Inghilleri, M., Liberati, G., Mattia, D., opinions of the utility of the system, generally in Pizzimenti, A., Riccio, A., Schettini, F., & Simione, line with their perceived performance at controlling L. (2015). Developing brain-computer interfaces the device. In conclusion, these two findings have from a user-centered perspective: Assessing the implications for the engineering design and clinical needs of persons with amyotrophic lateral sclero- care phases of assistive device deployment. sis, caregivers, and professionals. Applied Ergo- nomics, 50, 139-46. Page 21 Marchetti, M., & Priftis, K. (2015). Brain-computer to provide therapeutic benefits by inducing cortical interfaces in amyotrophic lateral sclerosis: A meta- reorganization via neuronal plasticity. This article nalysis. Clinical Neurophysiology, 126(6), 1255-63. presents a state-of-the-art review of BCI technology PMID: 25449558 used after nervous system injuries, specifically: amyo- ABSTRACT: OBJECTIVE: Despite recent ground- trophic lateral sclerosis, Parkinson’s disease, spinal breaking findings on the genetic causes of amyo- cord injury, stroke, and disorders of consciousness. trophic lateral sclerosis (ALS), and improvements Also presented is transcending, innovative research on neuroimaging techniques for ALS diagnosis have involving new treatment of neurological disorders. been reported, the main clinical intervention in ALS remains palliative care. Brain-computer interfaces Amico, E., Gómez, F., Habbal, D., Horki, P., Laureys, (BCIs) have been proposed as a channel of communi- S., Lebeau, M., Lesenfants, D., Lugo, Z., Müller-Putz, cation and control for ALS patients. The present meta- G., Noirhomme, Q., Pokorny, C., & Soddu, A. (2014). analysis was performed to test the evidence of BCI An independent SSVEP-based brain-computer effectiveness in ALS, and to investigate whether the interface in locked-in syndrome. Journal of Neural promising aims emerged from the first studies have Engineering, 11(3), 035002. been reached. METHODS: Studies on ALS patients PMID: 24838215 tested with BCIs, until June 2013, were searched in ABSTRACT: OBJECTIVE: Steady-state visually PubMed and PsychInfo. The random-effect approach evoked potential (SSVEP)-based brain-computer was used to compute the pooled effectiveness of BCI interfaces (BCIs) allow healthy subjects to commu- in ALS. A meta-regression was performed to test nicate. However, their dependence on gaze control whether there was a BCI performance improvement prevents their use with severely disabled patients. as a function of time. Finally, BCI effectiveness Gaze-independent SSVEP-BCIs have been designed for complete paralyzed ALS patients was tested. but have shown a drop in accuracy and have not been Twenty-seven studies were eligible for meta-analysis. tested in brain-injured patients. In the present paper, RESULTS: The pooled classification accuracy (C.A.) we propose a novel independent SSVEP-BCI based of ALS patients with BCI was about 70 percent, but on covert attention with an improved classification this estimation was affected by significant heteroge- rate. We study the influence of feature extraction algo- neity and inconsistency. C.A. did not significantly rithms and the number of harmonics. Finally, we test increase as a function of time. C.A. of completely online communication on healthy volunteers and pa- paralyzed ALS patients with BCI did not differ from tients with locked-in syndrome (LIS). APPROACH: that obtained by chance. CONCLUSIONS: After 15 Twenty-four healthy subjects and six LIS patients years of studies, it is as yet not possible to reliably participated in this study. An independent covert two- establish the effectiveness of BCIs. SIGNIFICANCE: class SSVEP paradigm was used with a newly devel- Methodological issues among the retrieved studies oped portable light emitting diode-based ‘interlaced should be addressed and new well-powered studies squares’ stimulation pattern. MAIN RESULTS: Mean should be conducted to confirm BCI effectiveness offline and online accuracies on healthy subjects were for ALS patients. respectively 85 ± 2 percent and 74 ± 13 percent, with eight out of twelve subjects succeeding to commu- 2014 nicate efficiently with 80 ± 9 percent accuracy. Two out of six LIS patients reached an offline accuracy Adeli, H., Buford, J.A., & Burns, A. (2014). Brain- above the chance level, illustrating a response to a computer interface after nervous system injury. command. One out of four LIS patients could com- Neuroscientist, 20(6), 639-51. municate online. SIGNIFICANCE: We have demon- PMID: 25193343 strated the feasibility of online communication with ABSTRACT: Brain-computer interface (BCI) has a covert SSVEP paradigm that is truly independent proven to be a useful tool for providing alternative of all neuromuscular functions. The potential clinical communication and mobility to patients suffering use of the presented BCI system as a diagnostic (i.e., from nervous system injury. BCI has been and will detecting command-following) and communication continue to be implemented into rehabilitation prac- tool for severely brain-injured patients will need to tices for more interactive and speedy neurological re- be further explored. covery. The most exciting BCI technology is evolving Page 22 Carmack, C.S., Mak, J.N., McCane, L.M., McFarland, standard communication pathways or other assistive D.J., Sellers, E.W., Vaughan, T.M.,Wolpaw, J.R., & technology. It has previously been shown that people Zeitlin, D. (2014). Brain-computer interface (BCI) with amyotrophic lateral sclerosis can successfully evaluation in people with amyotrophic lateral use BCI after all other means of independent com- sclerosis. Amyotrophic Lateral Sclerosis & Fronto- munication have failed. The BCI literature has as- temporal Degeneration, 15(3-4), 207-15. serted that brainstem stroke survivors can also benefit PMID: 24555843 from BCI use. This study used a P300-based event- Available in full-text at: http://www.ncbi.nlm.nih. related potential spelling system. This case study gov/pmc/articles/PMC4427912/pdf/nihms686954. demonstrates that an individual locked-in owing to pdf. brainstem stroke was able to use a noninvasive BCI ABSTRACT: Brain-computer interfaces (BCIs) to communicate volitional messages. Over a period might restore communication to people severely of 13 months, the participant was able to successfully disabled by amyotrophic lateral sclerosis (ALS) or operate the system during 40 of 62 recording sessions. other disorders. We sought to: (1) define a protocol He was able to accurately spell words provided by for determining whether a person with ALS can use a the experimenter and to initiate dialogues with his visual P300-based BCI, (2) determine what proportion family. The results broadly suggest that, regardless of this population can use the BCI, and (3) identify of the precipitating event, BCI use may be of benefit factors affecting BCI performance. Twenty-five indi- to those with locked-in syndrome. viduals with ALS completed an evaluation protocol using a standard 6 × 6 matrix and parameters selected Ikegami, S., Kansaku, K., Kondo, K., Saeki, N., & by stepwise linear discrimination. With an 8-channel Takano, K. (2014). A region-based two-step P300- EEG montage, the subjects fell into two groups in based brain-computer interface for patients with BCI accuracy (chance accuracy 3 percent). Seventeen amyotrophic lateral sclerosis. Clinical Neurophysi- averaged 92 (± 3) percent (range 71-100 percent), ology, 125(11), 2305-12. which is adequate for communication (G70 group). PMID: 24731767 Eight averaged 12 (± 6) percent (range 0-36 percent), ABSTRACT: OBJECTIVE: The P300-based brain- inadequate for communication (L40 subject group). computer interface (BCI) is designed to help patients Performance did not correlate with disability: 11/17 with motor disabilities to control their environment, (65 percent) of G70 subjects were severely disabled and it has been used successfully in patients with (i.e. ALSFRS-R < 5). All L40 subjects had visual im- amyotrophic lateral sclerosis (ALS). However, some pairments (e.g. nystagmus, diplopia, ptosis). P300 was ALS patients were unable to use the visual P300-BCI larger and more anterior in G70 subjects. A 16-chan- with the conventional row/column presentation. In nel montage did not significantly improve accuracy. this study, we evaluated the effect of a newly devel- In conclusion, most people severely disabled by ALS oped region-based two-step P300 speller, which has could use a visual P300-based BCI for communica- a larger flashing area than the conventional visual tion. In those who could not, visual impairment was array. METHODS: Seven ALS patients and seven the principal obstacle. For these individuals, auditory age- and sex-matched able-bodied control subjects P300-based BCIs might be effective. were required to input hiragana characters using our P300 BCI system. We prepared two types of input Hauser, C.K., Ryan, D.B., Sellers, E.W. (2014). procedures, the conventional row/column (RC) Noninvasive brain-computer interface enables speller and the two-step speller, and evaluated their communication after brainstem stroke. Science online performance. RESULTS: The mean online ac- Translational Medicine, 6(257), 257re7. curacy of the ALS patients was 24 percent for the RC PMID: 25298323 condition and 55 percent for the two-step condition. ABSTRACT: Brain-computer interfaces (BCIs) The accuracy of the control subjects was 71 and 83 provide communication that is independent of muscle percent for the RC and two-step condition, respec- control, and can be especially important for individu- tively. Accuracy in ALS patients was significantly als with severe neuromuscular disease who cannot use lower than that in the control subjects, and the new Page 23 visual stimuli significantly increased accuracy of ALS ALS does not affect the ability to control a BCI for patients. Using the new speller, two ALS patients communication. BCI performance can be maintained showed an initial accuracy sufficient for practical in the different stages of the illness. use (>70 percent). The other two ALS patients, who performed better in the first trial using the new speller, Breshears, J., Chang, E.F., & Rowland, N.C. (2013). continued to experience the BCI system, and their Neurosurgery and the dawning age of brain-ma- mean accuracy increased to 92 percent. CONCLU- chine interfaces. Surgical Neurology International, SIONS: The two-step procedure for the visual P300 4(Suppl 1), S11-4. BCI system provided significantly increased accuracy PMID: 23653884 for ALS patients compared with a conventional RC Available in full-text at: http://www.ncbi.nlm.nih. speller. SIGNIFICANCE: The new region-based gov/pmc/articles/PMC3642748. two-step P300 speller was effective in ALS patients, ABSTRACT: Brain-machine interfaces (BMIs) and the system may be beneficial to expand their are on the horizon for clinical neurosurgery. Elec- range of activities. trocorticography-based platforms are less invasive than implanted microelectrodes, however, the latter 2013 are unmatched in their ability to achieve fine motor control of a robotic prosthesis capable of natural hu- Birbaumer, N., Cavinato, M., Piccione, F., Ruf, C.A., man behaviors. These technologies will be crucial Silvoni, S., & Volpato, C. (2013). Amyotrophic to restoring neural function to a large population of lateral sclerosis progression and stability of brain- patients with severe neurologic impairment - includ- computer interface communication. Amyotrophic ing those with spinal cord injury, stroke, limb amputa- Lateral Sclerosis & Frontotemporal Degeneration, tion, and disabling neuromuscular disorders such as 14(5-6), 390-6. amyotrophic lateral sclerosis. On the opposite end of PMID: 23445258 the spectrum are neural enhancement technologies for ABSTRACT: Our objective was to investigate specialized applications such as combat. An ongoing the relationship between brain-computer interface ethical dialogue is imminent as we prepare for BMI (BCI) communication skill and disease progression platforms to enter the neurosurgical realm of clinical in amyotrophic lateral sclerosis (ALS). We sought management. also to assess stability of BCI communication performance over time and whether it is related to the Han, C.H., Hwang, H.J., Im. C.H., Jung, K.Y., & Lim, progression of neurological impairment before en- J.H. (2013). Classification of binary intentions tering the locked-in state. A three years follow-up, for individuals with impaired oculomotor func- BCI evaluation in a group of ALS patients (n = 24) tion: ‘eyes-closed’ SSVEP-based brain-computer was conducted. For a variety of reasons only three interface (BCI). Journal of Neural Engineering, patients completed the three years follow-up. BCI 10(2), 026021. communication skill and disability level, using the PMID: 23528484 Amyotrophic Lateral Sclerosis Functional Rating ABSTRACT: OBJECTIVE: Some patients suffering Scale-Revised, were assessed at admission and at from severe neuromuscular diseases have difficulty each of the three follow-ups. Multiple non-parametric controlling not only their bodies but also their eyes. statistical methods were used to ensure reliability of Since these patients have difficulty gazing at specific the dependent variables: correlations, paired test and visual stimuli or keeping their eyes open for a long factor analysis of variance. Results demonstrated no time, they are unable to use the typical steady-state vi- significant relationship between BCI communication sual evoked potential (SSVEP)-based brain-computer skill (BCI-CS) and disease evolution. The patients interface (BCI) systems. In this study, we introduce who performed the follow-up evaluations preserved a new paradigm for SSVEP-based BCI, which can their BCI-CS over time. Patients’ age at admission be potentially suitable for disabled individuals with correlated positively with the ability to achieve con- impaired oculomotor function. APPROACH: The trol over a BCI. In conclusion, disease evolution in proposed electroencephalography (EEG)-based BCI Page 24 system allows users to express their binary intentions Gomez-Gil, J., & Nicolas-Alonso, L.F. (2012). without needing to open their eyes. A pair of glasses Brain computer interfaces: A review. Sensors, with two light emitting diodes flickering at different 12(2), 1211-79. frequencies was used to present visual stimuli to PMID: 22438708 participants with their eyes closed, and we classified Available in full-text at: http://www.ncbi.nlm.nih.gov/ the recorded EEG patterns in the online experiments pmc/articles/PMC3304110/pdf/sensors-12-01211. conducted with five healthy participants and one pa- pdf. tient with severe amyotrophic lateral sclerosis (ALS). ABSTRACT: A brain-computer interface (BCI) is a MAIN RESULTS: Through offline experiments hardware and software communications system that performed with 11 participants, we confirmed that permits cerebral activity alone to control computers or human SSVEP could be modulated by visual selec- external devices. The immediate goal of BCI research tive attention to a specific light stimulus penetrating is to provide communications capabilities to severely through the eyelids. Furthermore, the recorded EEG patterns could be classified with accuracy high enough disabled people who are totally paralyzed or ‘locked for use in a practical BCI system. After customizing in’ by neurological neuromuscular disorders, such as the parameters of the proposed SSVEP-based BCI amyotrophic lateral sclerosis, brain stem stroke, or paradigm based on the offline analysis results, binary spinal cord injury. Here, we review the state-of-the- intentions of five healthy participants were classified art of BCIs, looking at the different steps that form in real time. The average information transfer rate of a standard BCI: signal acquisition, preprocessing or our online experiments reached 10.83 bits min(-1). signal enhancement, feature extraction, classification A preliminary online experiment conducted with an and the control interface. We discuss their advantages, ALS patient showed a classification accuracy of 80 drawbacks, and latest advances, and we survey the percent. SIGNIFICANCE: The results of our offline numerous technologies reported in the scientific and online experiments demonstrated the feasibility literature to design each step of a BCI. First, the re- of our proposed SSVEP-based BCI paradigm. It is view examines the neuroimaging modalities used in expected that our ‘eyes-closed’ SSVEP-based BCI the signal acquisition step, each of which monitors a system can be potentially used for communication different functional brain activity such as electrical, of disabled individuals with impaired oculomotor magnetic or metabolic activity. Second, the review function. discusses different electrophysiological control signals that determine user intentions, which can be 2012 detected in brain activity. Third, the review includes some techniques used in the signal enhancement step Birbaumer, N., Piccione, F., Silvoni, S., & Wildgru- to deal with the artifacts in the control signals and ber, M. (2012). Ideomotor silence: The case of improve the performance. Fourth, the review stud- complete paralysis and brain-computer interfaces ies some mathematic algorithms used in the feature (BCI). Psychological Research, 76(2), 183-91. PMID: 22252304 extraction and classification steps which translate ABSTRACT: The paper presents some specula- the information in the control signals into commands tions on the loss of voluntary responses and operant that operate a computer or other device. Finally, the learning in long-term paralysis in human patients review provides an overview of various BCI applica- and curarized rats. Based on a reformulation of the tions that control a range of devices. ideomotor thinking hypothesis already described in the 19th century, we present evidence that instru- Krusienski, D.J., Shih, J.J., & Wolpaw, J.R. (2012). mentally learned responses and intentional cognitive Brain-computer interfaces in medicine. Mayo processes extinguish as a consequence of long-term Clinic Proceedings, 87(3), 268-79. complete paralysis in patients with amyotrophic PMID: 22325364 lateral sclerosis (ALS). Preliminary data collected Available in full-text at: http://www.ncbi.nlm.nih. with ALS patients during extended and complete gov/pmc/articles/PMC3497935/pdf/main.pdf. paralysis suggest semantic classical conditioning of ABSTRACT: Brain-computer interfaces (BCIs) brain activity as the only remaining communication acquire brain signals, analyze them, and translate possibility in those states. them into commands that are relayed to output devices that carry out desired actions. BCIs do not use Page 25 normal neuromuscular output pathways. The main ALS regarding BCI design. This information will goal of BCI is to replace or restore useful function to guide BCIs in development to meet end-user needs. people disabled by neuromuscular disorders such as A telephone survey was undertaken of 61 people amyotrophic lateral sclerosis, cerebral palsy, stroke, with ALS from the University of Michigan’s Motor or spinal cord injury. From initial demonstrations of Neuron Disease Clinic. With regard to BCI design, electroencephalography-based spelling and single- participants prioritized accuracy of command iden- neuron-based device control, researchers have gone tification of at least 90 percent (satisfying 84 percent on to use electroencephalographic, intracortical, of respondents), speed of operation comparable to at electrocorticographic, and other brain signals for least 15-19 letters per minute (satisfying 72 percent), increasingly complex control of cursors, robotic and accidental exits from a standby mode not more arms, prostheses, wheelchairs, and other devices. than once every 2-4 hours (satisfying 84 percent). Brain-computer interfaces may also prove useful for While 84 percent of respondents would accept using rehabilitation after stroke and for other disorders. an electrode cap, 72 percent were willing to undergo In the future, they might augment the performance outpatient surgery and 41 percent to undergo surgery of surgeons or other medical professionals. Brain- with a short hospital stay in order to obtain a BCI. computer interface technology is the focus of a rapidly In conclusion, people with ALS expressed a strong growing research and development enterprise that is interest in obtaining BCIs, but current BCIs do not greatly exciting scientists, engineers, clinicians, and yet provide desired BCI performance. the public in general. Its future achievements will depend on advances in three crucial areas. Brain- 2010 computer interfaces need signal-acquisition hardware that is convenient, portable, safe, and able to function Arshak, K., Burke, E., Condron, J., & Kaneswaran, in all environments. Brain-computer interface systems K. (2010). Towards a brain controlled assistive need to be validated in long-term studies of real-world technology for powered mobility. Conference pro- use by people with severe disabilities, and effective ceedings: Annual International Conference of the and viable models for their widespread dissemination IEEE Engineering in Medicine and Biology Society, must be implemented. Finally, the day-to-day and 4176-80. moment-to-moment reliability of BCI performance PMID: 21096887 must be improved so that it approaches the reliability ABSTRACT: For individuals with mobility limita- of natural muscle-based function. tions, powered wheelchair systems provide improved functionality, increased access to healthcare, education 2011 and social activities. Input devices such as joystick and switches can provide the necessary input required Gruis, K.L., Huggins, J.E., & Wren, P.A. (2011). for efficient control of the powered wheelchair. For What would brain-computer interface users want? persons with limited dexterity, or fine control of the Opinions and priorities of potential users with fingers, access to mechanical hardware such as but- amyotrophic lateral sclerosis. Amyotrophic Lateral tons and joysticks can be quite difficult and sometimes Sclerosis, 12(5), 318-24. painful. For individuals with conditions such as trau- PMID: 21534845 matic brain injury, multiple sclerosis or amyotrophic Available in full-text at: http://www.ncbi.nlm.nih. lateral sclerosis voluntary control of limb movement gov/pmc/articles/PMC3286341/pdf/nihms-350696. maybe substantially limited or completely absent. pdf. Brain Computer Interfaces (BCI) are emerging as a ABSTRACT: Universal design principles advocate possible method to replace the brains normal output inclusion of end users in every design stage, including pathways of peripheral nerves and muscles, allowing research and development. Brain-computer interfaces individuals with paralysis a method of communication (BCIs) have long been described as potential tools and computer control. This study involves the analysis to enable people with amyotrophic lateral sclero- of non-invasive electroencephalograms (EEG) arising sis (ALS) to operate technology without moving. from the use of a newly developed Human Machine Therefore the objective of the current study is to Interface (HMI) for powered wheelchair control. determine the opinions and priorities of people with Using a delayed response task, binary classification Page 26 of left and right movement intentions were classified and fatigue might be reduced during operation of a with a best classification rate of 81.63 percent from BCI associated with human natural motor behavior. single trial EEG. Results suggest that this method may SIGNIFICANCE: The development of a user-friendly be used to enhance control of HMI’s for individuals BCI will promote practical BCI applications in para- with severe mobility limitations. lyzed patients.

Bai, O., Fei, D.Y., Floeter, M.K., Huang, D., & Lin, P. Bensch, M., Halder, S., Kubler, A., Mugler, E.M., & (2010). Towards a user-friendly brain-computer Ruf, C.A. (2010). Design and implementation of interface: Initial tests in ALS and PLS patients. a P300-based brain-computer interface for con- Clinical Neurophysiology, 121(8), 1293-303. trolling an internet browser. IEEE Transactions PMID: 20347612 on Neural Systems and Rehabilitation Engineering, Available in full-text at: http://www.ncbi.nlm.nih. 18(6), 599-609. gov/pmc/articles/PMC2895010/pdf/nihms-185609. PMID: 20805058 pdf. ABSTRACT: An electroencephalographic (EEG) ABSTRACT: OBJECTIVE: Patients usually require brain-computer interface (BCI) internet browser was long-term training for effective EEG-based brain- designed and evaluated with 10 healthy volunteers computer interface (BCI) control due to fatigue and three individuals with advanced amyotrophic caused by the demands for focused attention during lateral sclerosis (ALS), all of whom were given tasks prolonged BCI operation. We intended to develop a to execute on the internet using the browser. Partici- user-friendly BCI requiring minimal training and less pants with ALS achieved an average accuracy of 73 mental load. METHODS: Testing of BCI performance percent and a subsequent information transfer rate was investigated in three patients with amyotrophic (ITR) of 8.6 bits/min and healthy participants with lateral sclerosis (ALS) and three patients with primary no prior BCI experience over 90 percent accuracy lateral sclerosis (PLS), who had no previous BCI and an ITR of 14.4 bits/min. We define additional experience. All patients performed binary control criteria for unrestricted internet access for evalua- of cursor movement. One ALS patient and one PLS patient performed four-directional cursor control in a tion of the presented and future internet browsers, two-dimensional domain under a BCI paradigm asso- and we provide a review of the existing browsers ciated with human natural motor behavior using motor in the literature. The P300-based browser provides execution and motor imagery. Subjects practiced for unrestricted access and enables free web surfing for 5-10min and then participated in a multi-session study individuals with paralysis. of either binary control or four-directional control including online BCI game over 1.5-2h in a single Birbaumer, N., Escolano, C., Matuz, T., Minguez, J., visit. RESULTS: Event-related desynchronization & Ramos, M.A. (2010). A telepresence robotic sys- and event-related synchronization in the beta band tem operated with a P300-based brain-computer were observed in all patients during the production interface: Initial tests with ALS patients. Confer- of voluntary movement either by motor execution or ence Proceedings: Annual International Conference motor imagery. The online binary control of cursor of the IEEE Engineering in Medicine and Biology movement was achieved with an average accuracy Society, 2010, 4476-80. about 82.1+/-8.2 percent with motor execution and PMID: 21095775 about 80 percent with motor imagery, whereas offline ABSTRACT: Brain-computer interfaces (BCIs) open accuracy was achieved with 91.4+/-3.4 percent with a new valuable communication channel for people motor execution and 83.3+/-8.9 percent with motor with severe neurological or motor degenerative imagery after optimization. In addition, four-direc- diseases, such as ALS patients. On the other hand, tional cursor control was achieved with an accuracy the ability to teleoperate robots in a remote scenario of 50-60 percent with motor execution and motor provides a physical entity embodied in a real en- imagery. CONCLUSION: Patients with ALS or PLS vironment ready to perceive, explore, and interact. may achieve BCI control without extended training, The combination of both functionalities provides a Page 27 system with benefits for ALS patients in the context Sellers, E.W., Vaughan, T.M., & Wolpaw, J.R. (2010). of neurorehabilitation or maintainment of the neural A brain-computer interface for long-term inde- activity. This paper reports a BCI telepresence system pendent home use. Amyotrophic Lateral Sclerosis, which offers navigation, exploration and bidirectional 11(5), 449-55. communication, only controlled by brain activity; and PMID: 20583947 an initial study of applicability with ALS patients. ABSTRACT: Our objective was to develop and The results show the feasibility of this technology in validate a new brain-computer interface (BCI) sys- real patients. tem suitable for long-term independent home use by people with severe motor disabilities. The BCI was Furdea, A., Halder, S., Hösle, A., Kleih, S.C., Kübler, used by a 51-year-old male with ALS who could no A., Münßinger, J.I., & Raco, V. (2010). Brain paint- longer use conventional assistive devices. Caregivers ing: First evaluation of a new brain-computer in- learned to place the electrode cap, add electrode gel, terface application with ALS-patients and healthy and turn on the BCI. After calibration, the system volunteers. Frontiers in Neuroscience, 4, 182. allowed the user to communicate via EEG. Re-cali- PMID: 21151375 bration was performed remotely (via the internet), and BCI accuracy assessed in periodic tests. Reports of Available in full-text at: http://www.ncbi.nlm.nih. BCI usefulness by the user and the family were also gov/pmc/articles/PMC2996245. recorded. Results showed that BCI accuracy remained ABSTRACT: Brain-computer interfaces (BCIs) en- at 83 percent (r = -.07, n.s.) for over 2.5 years (1.4 able paralyzed patients to communicate; however, percent expected by chance). The BCI user and his up to date, no creative expression was possible. family state that the BCI had restored his indepen- The current study investigated the accuracy and dence in social interactions and at work. He uses the user-friendliness of P300-Brain Painting, a new BCI BCI to run his NIH-funded research laboratory and application developed to paint pictures using brain ac- to communicate via e-mail with family, friends, and tivity only. Two different versions of the P300-Brain colleagues. In addition to this first user, several other Painting application were tested: A colored matrix similarly disabled people are now using the BCI in tested by a group of ALS-patients (n = 3) and healthy their daily lives. In conclusion, long-term independent participants (n = 10), and a black and white matrix home use of this BCI system is practical for severely tested by healthy participants (n = 10). The three disabled people, and can contribute significantly to ALS-patients achieved high accuracies; two of them quality of life and productivity. reaching above 89 percent accuracy. In healthy subjects, a comparison between the P300-Brain Painting 2009 application (colored matrix) and the P300-Spelling application revealed significantly lower accuracy and Furdea, A., Halder, S., Hammer, E.M., Kotchoubey, P300 amplitudes for the P300-Brain Painting appli- B., Kübler, A., & Nijboer, F. (2009). A brain-com- cation. This drop in accuracy and P300 amplitudes puter interface controlled auditory event-related was not found when comparing the P300-Spelling potential (p300) spelling system for locked-in pa- application to an adapted, black and white matrix of tients. Annals of the New York Academy of Sciences, the P300-Brain Painting application. By employing 1157, 90-100. a black and white matrix, the accuracy of the P300- PMID: 19351359 ABSTRACT: Using brain-computer interfaces Brain Painting application was significantly enhanced (BCI) humans can select letters or other targets on a and reached the accuracy of the P300-Spelling ap- computer screen without any muscular involvement. plication. ALS-patients greatly enjoyed P300-Brain An intensively investigated kind of BCI is based on Painting and were able to use the application with the the recording of visual event-related brain potentials same accuracy as healthy subjects. P300-Brain Paint- (ERP). However, some severely paralyzed patients ing enables paralyzed patients to express themselves who need a BCI for communication have impaired creatively and to participate in the prolific society vision or lack control of gaze movement, thus mak- through exhibitions. ing a BCI depending on visual input no longer fea- Page 28 sible. In an effort to render the ERP-BCI usable for lateral sclerosis or stroke can transmit up to 80 bits/ this group of patients, the ERP-BCI was adapted to min of information, the use of BCIs - invasive or auditory stimulation. Letters of the alphabet were as- noninvasive - in severely or totally paralyzed patients signed to cells in a 5 x 5 matrix. Rows of the matrix has met some unforeseen difficulties. SUMMARY: were coded with numbers 1 to 5, and columns with Invasive and noninvasive BCIs using recordings from numbers 6 to 10, and the numbers were presented nerve cells, large neuronal pools such as electrocor- auditorily. To select a letter, users had to first select ticogram and electroencephalography, or blood flow the row and then the column containing the desired based measures such as functional magnetic reso- letter. Four severely paralyzed patients in the end- nance imaging and near-infrared spectroscopy show stage of a neurodegenerative disease were examined. potential for communication in locked-in syndrome All patients performed above chance level. Spelling and movement restoration in chronic stroke, but con- accuracy was significantly lower with the auditory trolled phase III clinical trials with larger populations system as compared with a similar visual system. Pa- of severely disturbed patients are urgently needed. tients reported difficulties in concentrating on the task when presented with the auditory system. In future Birbaumer, N., Furdea, A., Halder, S., Jordan, M.A., studies, the auditory ERP-BCI should be adjusted by Krusienski, D.J., Kübler, A., Matuz, T., Mellinger, taking into consideration specific features of severely J., Mochty, U., Nijboer, F., Sellers, E.W., Vaughan, paralyzed patients, such as reduced attention span. T.M., & Wolpaw, J.R. (2008). A P300-based brain- This adjustment in combination with more intensive computer interface for people with amyotrophic training will show whether an auditory ERP-BCI can lateral sclerosis. Clinical Neurophysiology, 119(8), become an option for visually impaired patients. 1909-16. PMID: 18571984 2008 Available in full-text at: http://www.ncbi.nlm.nih. gov/pmc/articles/PMC2853977/pdf/nihms59800.pdf. Birbaumer, N., Cohen, L., & Murguialday, A.R. ABSTRACT: OBJECTIVE: The current study (2008). Brain-computer interface in paralysis. evaluates the efficacy of a P300-based brain-computer Current Opinion in Neurology, 21(6), 634-8. interface (BCI) communication device for individu- PMID: 18989104 als with advanced ALS. METHODS: Participants ABSTRACT: PURPOSE OF REVIEW: Communica- attended to one cell of a N x N matrix while the N tion with patients suffering from locked-in syndrome rows and N columns flashed randomly. Each cell and other forms of paralysis is an unsolved challenge. of the matrix contained one character. Every flash Movement restoration for patients with chronic stroke of an attended character served as a rare event in or other brain damage also remains a therapeutic prob- an oddball sequence and elicited a P300 response. lem and available treatments do not offer significant Classification coefficients derived using a stepwise improvements. This review considers recent research linear discriminant function were applied to the data in brain-computer interfaces (BCIs) as promising after each set of flashes. The character receiving the solutions to these challenges. RECENT FINDINGS: highest discriminant score was presented as feedback. Experimentation with nonhuman primates suggests RESULTS: In Phase I, six participants used a 6 x 6 that intentional goal directed movements of the up- matrix on 12 separate days with a mean rate of 1.2 per limbs can be reconstructed and transmitted to selections/min and mean online and offline accuracies external manipulandum or robotic devices controlled of 62 percent and 82 percent, respectively. In Phase II, from a relatively small number of microelectrodes four participants used either a 6 x 6 or a 7 x 7 matrix implanted into movement-relevant brain areas after to produce novel and spontaneous statements with some training, opening the door for the develop- a mean online rate of 2.1 selections/min and online ment of BCI or brain-machine interfaces in humans. accuracy of 79 percent. The amplitude and latency of Although noninvasive BCIs using electroencephalo- the P300 remained stable over 40 weeks. CONCLU- graphic recordings or event-related-brain-potentials SIONS: Participants could communicate with the in healthy individuals and patients with amyotrophic P300-based BCI and performance was stable over Page 29 many months. SIGNIFICANCE: BCIs could provide neuromuscular disorders and may eventually become an alternative communication and control technology useful to less severely disabled and/or healthy in- in the daily lives of people severely disabled by ALS. dividuals across a wide range of applications. This review discusses the structure and functions of BCI Daly, J.J., & Wolpaw, J.R. (2008). Brain-computer systems, clarifies terminology and addresses practical interfaces in neurological rehabilitation. The Lan- applications. Progress and opportunities in the field cet: Neurology, 7(11), 1032-43. are also identified and explicated. PMID: 18835541 ABSTRACT: Recent advances in analysis of brain 2006 signals, training patients to control these signals, and improved computing capabilities have enabled Birbaumer, N., Flor, H., Hinterberger, T., Karim, A.A., people with severe motor disabilities to use their brain Kübler, A., Mellinger, J., Neumann, N., & Richter, J. signals for communication and control of objects in (2006). Neural internet: Web surfing with brain their environment, thereby bypassing their impaired potentials for the completely paralyzed. Neurore- neuromuscular system. Non-invasive, electroen- habilitation and Neural Repair, 20(4), 508-15. cephalogram (EEG)-based brain-computer interface PMID: 17082507 (BCI) technologies can be used to control a computer ABSTRACT: Neural Internet is a new technological cursor or a limb orthosis, for word processing and advancement in brain-computer interface research, accessing the internet, and for other functions such which enables locked-in patients to operate a Web as environmental control or entertainment. By re- browser directly with their brain potentials. Neural establishing some independence, BCI technologies Internet was successfully tested with a locked-in can substantially improve the lives of people with patient diagnosed with amyotrophic lateral sclero- devastating neurological disorders such as advanced sis rendering him the first paralyzed person to surf amyotrophic lateral sclerosis. BCI technology might the Internet solely by regulating his electrical brain also restore more effective motor control to people activity. The functioning of Neural Internet and its after stroke or other traumatic brain disorders by clinical implications for motor-impaired patients are helping to guide activity-dependent brain plasticity highlighted. by use of EEG brain signals to indicate to the patient the current state of brain activity and to enable the 2005 user to subsequently lower abnormal activity. Alterna- tively, by use of brain signals to supplement impaired Birbaumer, N., Kübler, A., McFarland, D.J., Mel- muscle control, BCIs might increase the efficacy of linger, J., Nijboer, F., Pawelzik, H., Schalk, G., a rehabilitation protocol and thus improve muscle Vaughan, T.M., & Wolpaw, J.R. (2005). Patients control for the patient. with ALS can use sensorimotor rhythms to operate a brain-computer interface. Neurology, 64(10), 2007 1775-7. PMID: 15911809 Allison, B.Z., Wolpaw, E.W., & Wolpaw, J.R. (2007). ABSTRACT: People with severe motor disabilities Brain-computer interface systems: progress and can maintain an acceptable quality of life if they can prospects. Expert Review of Medical Devices, 4(4), communicate. Brain-computer interfaces (BCIs), 463-74. which do not depend on muscle control, can provide PMID: 17605682 communication. Four people severely disabled by ABSTRACT: Brain-computer interface (BCI) sys- ALS learned to operate a BCI with EEG rhythms tems support communication through direct measures recorded over sensorimotor cortex. These results of neural activity without muscle activity. BCIs may suggest that a sensorimotor rhythm-based BCI could provide the best and sometimes the only communi- help maintain quality of life for people with ALS. cation option for users disabled by the most severe Page 30 2004 PMID: 12899265 ABSTRACT: Training severely paralyzed patients Ostrosky-Solís, F., Ramírez, M., & Santana, D. to use a brain-computer interface (BCI) for com- (2004). [Recent advances in rehabilitation tech- munication poses a number of issues and problems. nology: a review of the brain-computer interface]. Over the past six years, we have trained 11 patients Revista de Neurologia, 39(5), 447-50. [Article in to self-regulate their slow cortical brain potentials Spanish]. and to use this skill to move a cursor on a computer PMID: 15378459 screen. This paper describes our experiences with this Available in full-text: http://www.revneurol.com/sec/ patient group including the problems of accepting and resumen.php?or=pubmed&id=2004117. rejecting patients, communicating and interacting PMID: 15378459 with patients, how training may be affected by social, ABSTRACT: INTRODUCTION AND AIMS: In familial, and institutional circumstances, and the this work we review some of the options available in importance of motivation and available reinforcers. rehabilitation technology that are used to aid people with severe neuromuscular disorders, and which take 2002 electrophysiological activity as a source of biological signals with which to design interfaces. DEVEL- Birbaumer, N., McFarland, D.J., Pfurtscheller, G., OPMENT: A number of different researchers have Vaughan, T.M., & Wolpaw, J.R. (2002). Brain-com- generated a novel communication and control system puter interfaces for communication and control. that utilizes the electrical activity of the brain as a Clinical Neurophysiology, 113(6), 767-91. signal that represents the messages or commands an PMID: 12048038 individual sends to the outside world, without using ABSTRACT: For many years people have specu- the normal output pathways of the brain, such as pe- lated that electroencephalographic activity or other ripheral nerves and muscles; instead, this is achieved electrophysiological measures of brain function might through an artificial system that extracts, encodes provide a new non-muscular channel for sending and applies them, called a brain-computer interface messages and commands to the external world - a (BCI). The electrophysiological activity for a BCI brain-computer interface (BCI). Over the past 15 can be obtained by means of superficial or implanted years, productive BCI research programs have arisen. electrodes, and may therefore be classified as invasive Encouraged by new understanding of brain function, or non-invasive. Five types of brain signals have been by the advent of powerful low-cost computer equip- explored for use with a BCI: visual evoked potentials, ment, and by growing recognition of the needs and slow cortical potentials, cortical neuronal activity, potentials of people with disabilities, these programs beta and mu rhythms, and event-related potentials. concentrate on developing new augmentative com- CONCLUSIONS: Thanks to recent improvements munication and control technology for those with and developments in prototypes, this technology is severe neuromuscular disorders, such as amyotrophic sure to open up new possibilities of communication lateral sclerosis, brainstem stroke, and spinal cord and control for the affected population; it also rep- injury. The immediate goal is to provide these users, resents a valuable field of multidisciplinary research who may be completely paralyzed, or ‘locked in’, with numerous interesting applications in areas be- with basic communication capabilities so that they yond the sphere of health care. can express their wishes to caregivers or even operate word processing programs or neuroprostheses. 2003 Present-day BCIs determine the intent of the user from a variety of different electrophysiological sig- Kübler, A., & Neumann, N. (2003). Training nals. These signals include slow cortical potentials, locked-in patients: A challenge for the use of P300 potentials, and mu or beta rhythms recorded brain-computer interfaces. IEEE Transactions from the scalp, and cortical neuronal activity recorded on Neural Systems and Rehabilitation Engineering, by implanted electrodes. They are translated in real- 11(2), 169-72. time into commands that operate a computer display Page 31 or other device. Successful operation requires that the user encode commands in these signals and that the BCI derive the commands from the signals. Thus, the user and the BCI system need to adapt to each other both initially and continually so as to ensure stable performance. Current BCIs have maximum information transfer rates up to 10-25bits/min. This limited capacity can be valuable for people whose severe disabilities prevent them from using conventional augmentative communication methods. At the same time, many possible applications of BCI technology, such as neuroprosthesis control, may require higher information transfer rates. Future progress will depend on: recognition that BCI research and development is an interdisciplinary problem, involving neurobiology, psychology, engineering, mathematics, and computer science; identification of those signals, whether evoked potentials, spontaneous rhythms, or neuronal firing rates, that users are best able to control independent of activity in conventional motor output pathways; development of training methods for helping users to gain and maintain that control; delineation of the best algorithms for translating these signals into device commands; attention to the identification and elimination of artifacts such as electromyographic and electro-oculographic activity; adoption of precise and objective procedures for evaluating BCI performance; recognition of the need for long-term as well as short- term assessment of BCI performance; identification of appropriate BCI applications and appropriate match- ing of applications and users; and attention to factors that affect user acceptance of augmentative technology, including ease of use, cosmesis, and provision of those communication and control capacities that are most important to the user. Development of BCI technology will also benefit from greater emphasis on peer-reviewed research publications and avoidance of the hyperbolic and often misleading media attention that tends to generate unrealistic expectations in the public and skepticism in other researchers. With adequate recognition and effective engagement of all these issues, BCI systems could eventually provide an important new communication and control option for those with motor disabilities and might also give those without disabilities a supplementary control channel or a control channel useful in special circumstances. Page 32 The Brief History of Brain Computer Interfaces www.brainvision.co.uk/blog/2014/04/the-brief- Quick Looks history-of-brain-computer-interfaces Christopher and Dana Reeve Paralysis Resource Center (PRC) PRC is promoting the health and well-being of people living with paralysis and their families through comprehensive infor- Online Resources Related to mation resources and referral services. Brain Computer Interface & Toll Free: 800/539-7309 Neuromuscular Diseases/Disorders Information Specialists are available from 9 a.m. until 8 p.m. EST (English/Spanish) Contact: feedback.paralysis.org/feedback/stage_1. Brain computer interface (BCI), also known as Free Comprehensive guide to paralysis-related brain- or mind-machine interface, is a direct com- topics in English and Spanish: http://tinyurl.com/ munication interface system between an external lxnaxgg. device and the brain, which allows an individual www.paralysis.org. to communicate with or control a computer or other electronic device using his or her brainwaves ClinicalTrials.gov without movement from the neuromuscular sys- A database that provides information about fed- tem. Individuals with neuromuscular diseases/ erally and privately supported clinical research disorders (NMD) such as amyotrophic lateral in human volunteers. Information provided in- sclerosis (ALS), locked-in syndrome, stiff person cludes the trial’s purpose, who may participate, syndrome, multiple sclerosis, etc. may benefit from locations and phone numbers for more details. BCI technologies. For more information about the Studies related to BCI: tinyurl.com/pd45jjk. use of the BCI as well as NMD we recommend the Studies related to NMD: http://tinyurl.com/ following resources: pm2g3do. www.clinicaltrials.gov. ALS Association (ALSA) Toll Free: 800/782-4747, 202/407-8580 How Brain-computer Interfaces Work Contact ALSA: computer.howstuffworks.com/brain-computer- www.alsa.org/about-us/contact-us.html. interface.htm Locate a Local Chapter: www.alsa.org/community/chapters. Locked-In Syndrome - Rehabilitation Institute FYI: Brain-Computer Interface (BCI) Informa- of Chicago Treatment Program tion Sheet – www.alsa.org/als-care/resources/ Phone: 312/238-1000 publications-videos/factsheets/brain-computer- lifecenter.ric.org/index.php?tray=content interface.html. &cid=3209. www.alsa.org. Locked-In Syndrome Information Page from American Parkinson Disease Association the National Institute of Neurological Disorders (APDA) and Stroke (NINDS) Toll Free: 800/223-2732, 718/981-8001 www.ninds.nih.gov/disorders/lockedinsyndrome/ Email: [email protected]. lockedinsyndrome.htm. Locate a Local Resources: Search other conditions: www.ninds.nih.gov/ www.apdaparkinson.org/local-resources. index.htm. www.apdaparkinson.org. Page 33 The Michael J. Fox Foundation National Spinal Cord Injury Association Toll Free: 800/708-7644 (NSCIA) michaeljfox.org. The National Spinal Cord Injury Association, a program of the United Spinal Association, serves Multiple Sclerosis Foundation (MSF) as a comprehensive information source for anyone Toll Free MS Helpline: 888/679-6287, Available effected by spinal cord injury. 9:00 a.m. until 7:00 p.m. EST Toll Free: 800/962-9629 Phone: 954/776-6805 Email: [email protected]. Email: [email protected]. Locate a Local Chapter: Locate a Local Support Group: www.spinalcord.org/chapters/directory. www.msfocus.org/support-groups.aspx. www.spinalcord.org. www.msfocus.org. The Pass It On Center: A National Collabora- Muscular Dystrophy Association (MDA) tion for the Reutilization and Coordination of Toll Free: 800/572-1717 Assistive Technology Email: [email protected]. Toll Free: 800/497-8665 Locate a Local Chapter: www.mdausa.org/locate. Email: [email protected]. What Is a Neuromuscular Disease? Information Reuse Locations: Guide: www.mda.org/publications/teachers-guide/ www.passitoncenter.org/reuse_locations.php. what-is-a-neuromuscular-disease. www.mdausa.org.

National Assistive Technology Technical Assis- tance Partnership (NATTAP) State Contact List: www.resnaprojects.org/allcontacts/statewidecontacts.html. State AT Financing Programs: www.resnaprojects. org/allcontacts/allafpcontacts.html.

National Institute of Neurological Disorders and Stroke (NINDS) Toll Free: 800/352-9424, 301/496-5751 (V) www.ninds.nih.gov.

National Multiple Sclerosis Society (NMSS) Toll Free: 800/344-4867 Locate a Local Chapter: www.nationalmssociety. org/find-a-chapter/index.aspx. www.nationalmssociety.org.

National Parkinson Foundation, Inc. (NPF) Toll Free: 800/473-4636 Email: [email protected]. Locate Local NPF Centers & Chapters: www.parkinson.org/Search-Pages/Chapter-Loca- tor. www.parkinson.org. Page 34 Search Terms for the BCI/NMD  Adaptation  Quadriplegia  Amyotrophic lateral sclerosis (ALS)  Qualitative Analysis  Art Therapy  Quality Of Life  Assistive Technology  Rare Disorders  Attitudes  Rehabilitation/Services/Technology  Augmentative & Alternative Communication  Robotics  Barriers  Sensory Impairments  Biofeedback  Service Utilization  Brain  Severe Disabilities  Paralysis  Software Systems  Case Studies  Speech Disorders/Impairments  Cognition  Spinal Cord Injuries  Coma  Stroke  Communication/Aids/Devices/Skills  Task Analysis  Computer(s)/Applications  Technology  Control Systems  Traumatic Brain Injury  Daily living  Universal Design  Degenerative Diseases  Wheeled Mobility  Devices  Electroencephalography (EEG)  Electrophysiology  Environmental Control Systems  Evaluation  Eye Movements  Feasibility Studies  Feedback  Home-Based  Independent living  Limbs  Locked-In  Man-Machine Systems  Mobility Impairments  Motor Skills  Movement Disorders  Multiple Sclerosis  Nerves  Neurofeedback  Neuroimaging  Neurological Impairments  Neuromuscular Diseases/Disorders  Paralysis  Parkinson’s Disease  Physical Disabilities  Prostheses and Implants About reSearch:

reSearch is an information product from the National Rehabilitation Information Center (NARIC). Each issue is based on real-world queries received by our information specialists from researchers, educators, and rehabilitation professionals around the world.

We search several sources both in-house and online, to fill these requests including:

 REHABDATA and the NIDRR Program database

 Education Resources Information Center

 National Clearinghouse of Rehabilitation Training Materials

 Campbell and Cochrane Collaborations

 PubMed and other National Library of Medicine databases

 Agency for Health Care Policy and Research databases

 Center for International Rehabilitation Research Information and Exchange

 and other reputable, scholarly information resources.

We hope you find thesereSearch briefs informative in your own research.

- NARIC Information and Media Team

NARIC is operated by HeiTech Serivces, Inc., for the National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR) under contract number GS-06F-0726Z.