SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx ·...

29
BCI Meeting Workshops SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00) WS 1: BCIs for stroke rehabilitation Presenters Christoph Guger, g.tec medical engineering GmbH, Austria José del R. Millán, Institute of Bioengineering, EPFL, Switzerland Vivek Prabhakaran, University of Wisconsin-Madison Radiology WIMR, USA Kyousuke Kamada, Asahikawa University, Japan Tetsuo Ota, Asahikawa University, Japan Milena Korostenskaja, Florida Hospital for Children, USA Michael Tangermann, University Freiburg, Germany David Lin, MGH Harvard, USA Abstract Lately, BCI systems become increasingly used in the context of stroke rehabilitation. Many BCI systems are based on motor imagery activity recorded from the sensorimotor cortex, which is translated into continuous control signals for rehabilitation devices. Some devices use Virtual Reality to allow users to observe an avatar’s limb movement. Other successful applications with patients use different brain stimulation techniques and/or robotic devices (such as exoskeletons or functional electrical stimulators) attached to patients’ paralyzed limbs or tongue stimulation. The workshop will review current stroke rehabilitation programs from different research labs and will provide insight into technology (EEG, MEG, fMRI), experimental setups (VR, FES, BCI), results and outcomes of patient studies in the acute, sub-acute or chronic state. Intended Audience - People interested in BCI technology, especially with motor imagery - People interested in clinical studies with stroke patients - People interested brain plasticity and brain stimulation Learning Objectives

Transcript of SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx ·...

Page 1: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

BCI Meeting Workshops

SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)

WS 1: BCIs for stroke rehabilitation

Presenters

Christoph Guger, g.tec medical engineering GmbH, AustriaJosé del R. Millán, Institute of Bioengineering, EPFL, SwitzerlandVivek Prabhakaran, University of Wisconsin-Madison Radiology WIMR, USAKyousuke Kamada, Asahikawa University, JapanTetsuo Ota, Asahikawa University, JapanMilena Korostenskaja, Florida Hospital for Children, USAMichael Tangermann, University Freiburg, GermanyDavid Lin, MGH Harvard, USA

Abstract

Lately, BCI systems become increasingly used in the context of stroke rehabilitation. Many BCI systems are based on motor imagery activity recorded from the sensorimotor cortex, which is translated into continuous control signals for rehabilitation devices. Some devices use Virtual Reality to allow users to observe an avatar’s limb movement. Other successful applications with patients use different brain stimulation techniques and/or robotic devices (such as exoskeletons or functional electrical stimulators) attached to patients’ paralyzed limbs or tongue stimulation.

The workshop will review current stroke rehabilitation programs from different research labs and will provide insight into technology (EEG, MEG, fMRI), experimental setups (VR, FES, BCI), results and outcomes of patient studies in the acute, sub-acute or chronic state.

Intended Audience

- People interested in BCI technology, especially with motor imagery- People interested in clinical studies with stroke patients- People interested brain plasticity and brain stimulation

Learning Objectives

1. Participants will learn about current technology for stroke rehabilitation2. Participants will be able to understand the target patient group3. Participants will learn about state-of-the art in BCI stroke rehabilitation

WS 6: Progress in Decoding Speech processes using intracranial signals

Presenters

Christian Herff, University of BremenTanja Schultz, University of BremenDean Krusienski, Old Dominion University

Page 2: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

Jon Brumberg, University of KansasPhil Kennedy, Neural SignalsTonio Ball, University of FreiburgEfraim Salari, UMC UtrechtJosh Chartier, UC San FranciscoJames O’Sullivan, Columbia UniversityStephanie Ries-Cornou, San Diego State UniversityBlaise Yvert, University Grenoble Alpes

Abstract

Speech provides a natural and efficient means of communication that is mostly unharnessed in current Brain-Computer Interfaces. Intracranial recordings allow for high spatial and temporal resolution recordings of cortical activity during speech process without the contamination by motion artifacts and thus enable in-depth analysis of the complex dynamics of speech processes.

In this workshop, we will discuss recent progress in processing and decoding speech using intracranial signals. The workshop will begin with short presentations describing work on speech synthesis, phonological context, auditory processing, articulatory representations, auditory attention, deep learning, communication disorders, among other topics. The subsequent discussion among all presenting experts will include future research directions and opportunities for collaboration and data sharing.

Intended Audience

This workshop is suitable for BCI researchers with biomedical, engineering and methodological backgrounds and will also include useful information for clinicians.

Learning Objectives

1. Participants will be able to identify limitations of non-invasive technologies for speech investigation.

2. Participants will be able to describe the state of the art in phone recognition in speech production and perception.

3. Participants will be able to define four milestones towards clinical use of speech prostheses.

WS 7: Noninvasive BCI-control of FES for grasp restoration in high spinal cord injured humans

Presenters

Gernot Müller-Putz, TU-GrazRüdiger Rupp, Heidelberg University HospitalAlexandra Vuckovic, University of GlasgowAndreea Sburlea, Graz BCI LabJoana Pereira, Graz University of Technology

Abstract

The bilateral loss of the grasp function associated with a complete lesion of the cervical spinal cord represents a severe handicap for social participation of the affected individuals. Any improvement of an impaired grasp function is highly desirable not only from the patient’s point of view but also for

Page 3: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

economic reasons. Motor neuroprostheses based on surface functional electrical stimulation (FES) provide a non-invasive option for generation of multiple grasp patterns. Closed-loop control of multi-pad FES electrodes allow for robust grasp pattern generation independent of wrist rotation.

EEG-based BCIs represent a valuable component of a neuroprosthetic user interface. A major advantage of BCIs is that they do not depend on residual motor functions. BCIs based on motor intention have enormous implications providing natural control of grasping neuroprostheses for individuals with a high spinal cord injury by relying on volitional signals recorded from the brain directly involved in upper extremity movements. This may also open up new avenues for using BCI-controlled neuroprosthesis as a tool in task-oriented, neurorestorative therapies.

This interdisciplinary workshop will give an update on the application of noninvasive BCIcontrolled grasp neuroprostheses in end users and will include a combination of coordinated invited talks together with demonstrations and structured targeted discussions.

Intended Audience

engineers, clinicians, researchers

Learning Objectives

1. Participants will understand the basic principles of noninvasive motor neuroprostheses2. Participants will understand the different concepts for a more intuitive BCI-control of

neuroprostheses3. Participants will have the ability for self-experiences of a BCI-controlled grasp

neuroprosthesis4. Participants will be aware of the challenges of the application of BCI-controlled

neuroprosthesis in end users with high spinal cord injury

WS 14: Collaborative and Competing Multi-Brain BCI’s

Presenters

Chris Berka, Advanced Brain Monitoring, Carlsbad, CA, USADavide Valeriani, University of Essex, ColchesterFabien Lotte, Inria, FranceJan van Erp, TNO, NetherlandsAnton Nijholt, University of Twente, NetherlandsTim Mullen, Intheon, San Diego, USA

Abstract

An emerging line of BCI research is the development of BCIs based on brain activity recorded from multiple users simultaneously. Applications include joint decision making in environments requiring high accuracy and/or rapid reactions or feedback; joint/shared control and movement planning of vehicles or robots; assess team performance, stress-aware task allocation, and rearrangement of tasks; characterization of group emotions, preferences, appreciations; social interaction research (two or more people); arts, entertainment, and games. Passive multi-brain research includes EEG hyperscanning to study social interaction, but also monitoring and integrating of brain activity to realize on-line improvement of group performance or online adaptation of a task or media display. In active multi-brain research we can distinguish between collaborative (movement planning, target detection) and competing multi-brain BCIs (e.g., in games). Various ways of merging and comparing

Page 4: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

brain activity of multiple users need to be considered and will be investigated in the workshop, this includes the computation of multi-users‘ features such as synchrony, models which are trained based on multi-users signals, and the development of BCI paradigms optimized for several users.

Intended Audience

This workshop is dedicated to BCI researchers, developers and users. It can also interest social neuroscientists and researchers in socio-affective computing.

Learning Objectives

1. Participants will be able to differentiate multi-users passive and active BCI2. Participants will be able to give at least two advantages of multi-users BCIs over traditional

BCIs3. Participant will be able to identify at least four challenges in multi-users BCI research (e.g.

synchronous data collection, BCI paradigms, features and models)4. Participants will be able to list three application domains for multi-user BCIs5. Participants will be able to select appropriate data processing methods for multi-brain data.6. Participants will be able to list at least two approaches to analyzing multi-brain recordings.

WS 15: ECoG based BCIs

Presenters

Gerwin Schalk, NCANDora Hermes, StanfordAyse Gunduz, Florida UniversityKai Miller, Stanford

Abstract

Electrocorticography (ECoG) is the technique of interacting with the brain electrically by stimulating or recording from the surface of the brain. ECoG has been used for decades for select clinical purposes – most commonly to identify functional and epileptic brain areas in people with epilepsy – and occasionally for research. The important role of ECoG for basic research and its potential to create a new range of clinical applications have long been under-appreciated.

Over the past several years, the unique qualities of ECoG have become widely and increasingly recognized by scientists engaged in basic and translational research. Basic research suggests that ECoG can elucidate brain function in ways that cannot be readily achieved using other imaging modalities, and translational research is producing exciting new ECoG-based applications that are already becoming available in the clinic.

Intended Audience

This program has been carefully designed to appeal to two target audiences. The program will be of interest to the scientist with an interest in theory and application of electrocorticographic (ECoG) signals recorded from the surface of the brain in humans or animals. The program will also have a strong appeal to neurologists, neurosurgeons, or clinical neurophysiologists who are interested in the clinical application of modern ECoG recording/stimulation technologies.

Learning Objectives

Page 5: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

1. Discuss the nature of brain signals recorded electrocorticographically (ECoG).2. Know about emerging understanding of ECoG physiology and of emerging techniques to

record it.3. Have an overview of current efforts in ECoG-based neuroscience.4. Contrast standard electrical brain stimulation and real-time functional ECoG mapping.5. Discuss the role of high frequency ECoG in functional assessment of brain activity.

WS 19: Examining Ethical Assumptions About Neural Engineering and BCI Development

Presenters

Paul Tubig, Philosophy, Center for Sensorimotor Neural Engineering, University of WashingtonJudy Illes, National Core for Neuroethics, University of British ColumbiaJonathan Wolpaw, Adaptive Center for Neurotechnology, Wadsworth Center, New YorkJane Huggins, Biomedical Engineering, University of MichiganLaura Specker Sullivan, Center for Bioethics, Harvard University

Abstract

The aim of this workshop is for participants to identify and examine the underlying ethical assumptions and values that shape the decisions and direction of BCI and neural engineering research. The workshop will be based on an ethical engagement approach developed by the Center for Sensorimotor Neural Engineering (CSNE) ethics thrust. The main workshop will consist of two activities: (1) a brief neuroethical survey to identify researchers’ ethical values related to BCI research, and (2) a facilitated dialogue with researchers about their answers on the survey. The general goal is to foster ethical reflexivity and to assure that the direction of BCI and neural engineering research accords with the ethical commitments the researchers themselves endorse. The learning objectives are the following: (1) understanding the importance of considering the ethical implications of BCI research, (2) identifying and examining the ethical assumptions and values shaping the design and conduct of BCI research, and (3) reflecting on and discussing how facilitated interdisciplinary dialogue can foster ethical reflexivity.

Intended Audience

Any stakeholder in the advancement of BCI and neural engineering research.

Learning Objectives

1. Participants learn about 3-5 reasons the importance of considering the ethical implications of BCI and neural engineering research. (Why ethics?)

2. Participants will identify and examine 4-5 of their own ethical assumptions and values that may implicitly or explicitly shape how they design and conduct BCI or neural engineering research. (What are your ethical assumptions?)

3. Participants will reflect on and discuss how facilitated interdisciplinary dialogue can foster ethical education and enhance the ethical framework guiding BCI and neural engineering researchers. (How can we best foster ethical reflection and practice?)

WS 22: Towards the Elusive Killer App for BCIs

Page 6: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

Presenters

Brendan AllisonAnjela VujicJing JinChristoph Guger

Abstract

Companies, universities, and other groups have been working toward a BCI “Killer App” for decades. Yet, despite major improvements such as wireless amplifiers, dry electrodes, and improved computing power, nobody has developed a BCI that appeals to most mainstream users. As implied by the title of this BCI Conference, BCIs may be getting lost in translation. Or, perhaps we simply don’t yet have the technology or ideas to inspire the BCI Killer App. This workshop will review some of the challenges in translating to mainstream (not clinical) populations. These include the need for more transparent, wearable sensors and new ways to incorporate the capabilities of modern sensing systems into mainstream apps. We will present new systems that strive toward addressing these goals, and participants will be able to try wearing them to see how they look and feel. The workshop will conclude with discussion, focused on development of new ideas or models to encourage further research.

Intended Audience

The audience does not require a background in any specific discipline. It should be of interest to a broad variety of experts who work with BCIs, or are interested in BCIs. Participants from academia, business, medicine, and other sectors are welcome.

Learning Objectives

1. Participants will learn about challenges in translating from laboratory to mainstream apps.2. Participants will learn about emerging solutions involving hardware design (including

wearable systems), interface development, and incorporation of emotion and user state.3. Participants will see examples of new systems through hands-on demonstrations and/or

videos, and thereby gain firsthand knowledge of the latest developments.

WS 23: User-Centered Design in BCI development; A Broad Perspective

Presenters

Prof. Dr. Andrea Kübler, University of WürzburgRay Grott, San Francisco State University and RESNA (Rehabilitation Engineering and Assistive Technology Society of North America)Katya Hill, University of PittsburghElmar Pels, University Medical Center UtrechtDr. Erik Aarnoutse, University Medical Center Utrecht

Abstract

In the last years BCIs have left the laboratory and came to the houses of the end-users. As daily use has become a reality, the feedback and remarks of users are valuable and cannot be overlooked. Therefore, to enhance the usability of BCIs and reduce the chance of assistive technology abandonment a user-centered design is pivotal. Simultaneously, we can learn from the perspective of

Page 7: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

technology design for more accessibility. In this workshop BCI researchers discuss their experiences with at home use and users’ feedback, and experts from outside the BCI field will share their insights about user-centered design. Additionally, participants to the workshop can send in questions to a person using a communication BCI at home for two years, the answers will be discussed at the workshop during an interactive and lively panel discussion.

Intended Audience

All BCI researchers working to bring their BCIs to the home of the end-user or researchers starting development of new BCIs to accomplish this goal.

Learning Objectives

1. Three principles of User-centered design.2. Three essential issues addressed by a user (user wishes).

Two misconceptions in BCI design; what researchers/clinicians think users want.

WS 25: Lower-limb brain-machine interfaces and their applications

Presenters

Kyuhwa Lee, Swiss Federal Institute of Technology in Lausanne (EPFL)Tomislav Milekovic, UNIGEAn Hong Do, University of California, IrvineJosé L. Contreras-Vidal, University of HoustonFabien Wagner, EPFL

Abstract

Lower-limb brain-machine interfaces (BMI) have been gaining more attention as its potential for assistive and clinical applications has become more prominent. Advances in robotic systems such as lower-limb exoskeletons and rehabilitative devices as well as more powerful machine learning algorithms have opened up new ideas and possibilities in such applications. More recently, rehabilitation coupled with BMI has been increasingly studied but still the majority of the effort has been put in upper limbs compared to lower limbs. Through this workshop, we intend to bring together researchers of different perspectives from neuroscience, robotics and data science to discuss about the current state of the lower-limb BMI and come up with a roadmap for future lower-limb BMI research.

Intended Audience

This workshop includes interdisciplinary topics such as decoding, neurorehabilitation and assistive technologies for lower-limb brain-machine interfaces. It is intended to bring researchers together in the fields of neuroscience, clinics, computer science, electronic engineering and robotics to discuss about effective ways to advance the current state of the art in lower-limb BMI research.

Learning Objectives

1. Share the state-of-the-art lower-limb BMI research.2. Identify current limitations of lower-limb BMI and possible solutions.3. Realistic roadmap of the protocol designs and evaluation methods for lower-limb BMI

research.

Page 8: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

SESSION 2- THURSDAY, MAY 24 (9:00-12:00

WS 2: BCIs for assessment of locked-in and DOC patients

Presenters

Christoph Guger, g.tec medical engineering GmbH, AustriaDamien Coyle, University of Ulster, UKDonatella Mattia, Neuroelectrical Imaging and BCI Lab, Fondazione Santa Lucia, ItalyChang S. Nam, North Carolina State University, USAJing Jin, East China University of science and Technology, ChinaKyousuke Kamada, Asahikawa Medical University, Japan

Abstract

Some patients diagnosed as vegetative are reclassified as (at least) minimally conscious when assessed by expert teams. A further subset of potentially communicative non-responsive patients might be undetectable through standard clinical testing. Other patients might have transient periods of relative wakefulness, but remain unaware of their surroundings. The workshop will provide an overview over groups that aim to use BCI technology to identify non-responsive patients or locked-in patients that might be able to communicate and use the technology as an assessment tool.

In the workshop recent experiments, analysis methods and results with EEG, fNIRS and fMRI will be shown and discussed. The goal of the workshop is to identify the most important trends of the last years and to facilitate interaction between participants.

Intended Audience

- People interested in Evoked Potentials (auditory and somatosensory)- People interested in BCI technology including motor imagery, P300, SSVEP- People interested in clinical studies with non-responsive patients and locked-in patients

Learning Objectives

1. Participants will learn about current technology for assessment and communication2. Participants will be able to understand the target patient group3. Participants learn about the state-of-the art in BCI coma assessment and communication

WS 4: Turning negative into positives! Exploiting “negative” results in Brain-Machine Interface research

Page 9: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

Presenters

Fabien Lotte, LaBRICamille Jeunet, EPFLLaurent Bougrain, LORIARicardo Chavarriaga, EPFLMoritz Grosse-Wentrup, Max Planck Institute for Intelligent Systems

Abstract

Scientific advancement is by essence a self-correcting process achieved through incremental confirmation or refusal of evolving hypothesis. Like other fields, progress in Brain-Machine interfacing (BMI) requires identifying whether a methodological approach is effective or not. However, unsuccessful attempts are barely reported in literature or discussed across research groups. This hampers BMI progress as 1) several groups may waste resources exploring the same methods unknowing it will not work, 2) it prevents the community from learning from its mistakes and from obtaining a bigger picture of BMI strengths and weaknesses. However, reporting these so-called “negative” results is not trivial, as their interpretation relies heavily on the way experiments are designed and conducted.

This workshop aims at discussing and promoting the reporting of relevant so-called “negative” results, from which we can draw significant conclusions. Besides inviting participants to report their own (unpublished) negative results, we will discuss criteria that help designing experiments and analysis criteria that allows extraction of useful knowledge, even when the initial hypothesis are not confirmed. Last but not least, we will advance means to promote the reporting of such results, e.g., by organizing a special issue about negative BMI research results in a BMI journal.

Intended Audience

Anyone involved in academic BCI/BMI research and wants to know more about socalled “negative results” in BCI/BMI research, what are those in our field, how they can be useful, and how to promote their reporting.

Learning Objectives

This workshop aims to convince the participants that the so-called “negative” results are actually results, and that their outcome can be positive, despite their misleading name. In particular:

1. Participants will learn at least 3 reasons why reporting on so-called negative results is important and even necessary for further progresses in BCI research

2. Participants will discover at least 4 examples of concrete useful negative results in BCI research

3. Participants will learn at least 3 criteria and principles to design BCI/BMI experiments maximizing the probability to obtain useful results, if they are negative.

4. Participants will have identified at least 3 criteria that make a negative results relevant and thus worth being published

WS 8: Eye Tracking, Vision, and BCI

Presenters

Page 10: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

Melanie Fried-Oken, PhD, CCC-SLP Bruce Wojciechowski, OD, FCOVDMichelle Kinsella, OTR/LBetts Peters, MA, CCC-SLP Brandon Eddy, MA, CCC-SLP Deniz Erdogmus, PhDLeslie Collins, PhDBoyla Mainsah, PhD

Abstract

Eye tracking and BCI, both individually and together in hybrid systems, have been used as assistive technology access methods for individuals with severe disabilities. Both methods are often used with visual interfaces, and thus require certain visual and oculomotor skills in potential users. This workshop will begin with an examination of the visual and oculomotor impairments often experienced by people with severe speech and physical disabilities, clinical presentation and assessment of these impairments, and barriers to eye tracking or BCI system use. Participants will then have an opportunity to experience simulated visual impairments and explore how they affect perception of visual BCI interfaces. After an overview of the use of eye tracking and BCI as assistive technology access methods, we will present results of studies involving modified visual interfaces and the fusion of EEG and eye tracking data, and discuss related engineering concepts. We will conclude with a group dialogue discussing recommendations and future research directions at the intersection of BCI, eye tracking, and visual impairment. An important topic of this discussion will be experimental and quantitative modeling approaches that can be followed to develop accurate models of the effect of visual impairments on brain and gaze signal acquisition.

Intended Audience

The intended audience for this workshop includes researchers and clinicians interested in BCI, eye tracking, and/or hybrid assistive technology systems for individuals with disabilities.

Learning Objectives

1. Participants will identify four visual impairments commonly found in people with severe speech and physical disabilities, and appropriate screening methods for each.

2. Participants will identify three interventions an eye care professional might use to optimize visual abilities for BCI or eye tracking users.

3. Participants will identify three potential applications of eye tracking or other vision-related data in BCI and assistive technology systems.

4. Participants will identify three system design modifications which may support effective use of BCI or eye tracking technology for users with SSPI.

5. Participants will identify three quantitative visual impairment modeling approaches which may support effective augmentation of signal analysis and intent classification algorithms for BCI or eye tracking technology users with SSPI.

WS 9: Natural Language Processing & BCI

Presenters

Steven Bedrick, Oregon Health and Science UniversityDavid Smith, Google

Page 11: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

Brian RoarkShiran Dudy, Oregon Health and Science University

Abstract

Natural Language Processing (NLP) is the computational analysis of spoken or written language. NLP plays an important role in many communications technologies, including speech recognition, spelling correction, and predictive typing systems. In AAC, the most common use of NLP is in the form of word completion and character and symbol suggestion. Such applications depend on n-gram and artificial neural network language models, and have the potential to be a crucial component of BCI systems. However, there exist substantial obstacles to translating NLP techniques to a BCI context. Abbreviations and other forms of shortening pose difficulties for traditional language models, and the inherently noisy channel of BCI means that entry errors are common. Furthermore, many modern techniques for language model training rely on large amounts of training data, which, when collected from users of BCI systems, poses substantial privacy concerns. The workshop organizers have practical, real-world experience with such issues (in both BCI and, crucially, in analogous non-BCI contexts such as mobile devices and medical records), and are excited to provide a venue for discussion. Additionally, the workshop will provide attendees with an overview of the fundamentals of NLP as applied to BCI, and is intended for a general scientific audience.

Intended Audience

While attendees with an engineering background will not be disappointed, the workshop will be accessible to any attendee of the meeting with some scientific or clinical background.

Learning Objectives

1. Participants will be able to explain what language models are in non-mathematical terms, and describe how they are used in BCI systems

2. Participants will be able to describe at least three ways that language modeling in a BCI context is different from traditional language modeling contexts

3. Participants will be able to describe two different approaches to securely and anonymously collecting private data for analysis

WS 11: BCI and Augmented/Virtual Reality

Presenters

Felix Putze, University BremenHakim Si Mohammed, IRISANick Waytowich, US Army ResearchSergi Bermudez i Badia and/or Athanasios Vourvopoulos, University MadeiraChristian Herff, University BremenJelena Mladenovic, Inria BordeauxTim Mullen, IntheonDean Krusienski, Old Dominion University

Abstract

VR/AR technology opens an exciting new field of research for the BCI community. BCI can be considered as a novel input device for VR/AR applications complementary to other modalities such as speech and gesture, capable of conveying the necessary information. Active BCIs notably allow a user

Page 12: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

to issue commands to the device or to enter text without physical involvement of any kind; passive BCIs monitor a user's state (e.g. workload level, attentional state) and can be used to proactively adapt the VR/AR interface. Additionally,

AR/VR offers the possibility for immersive scenarios in basic BCI research. To live up to these expectations, methodological advances are required for BCI interface and stimulus design, synchronization, or dealing with VR/AR specific artifacts and distractions. During the workshop, we will have a combination of presentations, live demonstrations, and discussions. Our goal is to create an understanding of the current capabilities of BCI in VR/AR applications, to create a taxonomy of current and future applications and to identify technical challenges. BCI researchers who are new to VR/AR will have the opportunity to learn about the necessary technical foundations.

Intended Audience

BCI researchers and practitioners at all levels of expertise. Both participants with focus on medical applications or general HCI settings. Experience with augmented or virtual reality is useful, but not required.

Learning Objectives

1. Participants will be able to identify four major areas within the taxonomy of applications of BCI for augmented and virtual reality scenarios.

2. Participants will know the technological foundations of and four state-of-the-art applications of BCI combined with augmented and virtual reality.

3. Participants will be able to name three major challenges of BCI specific to applications in AR/VR

WS 13: Recent Developments in Non-Invasive EEG SensorTechnology

Presenters

Chuck Anderson, Colorado State UniversityWalter Besio, University of Rhode IslandWalid Soussou, QuasarHasan Ayaz, Drexel University

Abstract

Conventional non-invasive EEG electrodes limit the quality of recorded EEG and BCI applications. Recently developed sensing technologies may lead to breakthroughs that are necessary for practical BCI. This workshop will discuss use, cost, and signal quality of sensing technologies such as no-prep dry electrodes, fNIRS, ultrasound, and tri-polar concentric electrodes.

Intended Audience

BCI researchers and practitioners who are curious about advances in sensing technology that may lead to more practical and reliable BCI applications.

Learning Objectives

1. Participants will be able to describe at least two limitations of current EEG sensing technology.

Page 13: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

2. Participants will be able to summarize the technology behind three new sensing developments.

3. Participants will be able to identify advantages and disadvantages of three new sensing technologies.

WS 17: Making use of the future of BCI implant technology

Presenters

Erik Aarnoutse, University Medical Center UtrechtTim Denison, MedtronicLuca Maiolo, CNR-IMMSamantha Santacruz, University of California, Berkeley

Abstract

Exciting developments are happening in implant neurotechnology and many BCI researchers can’t wait to use these technologies in their research. This workshop has prominent speakers from industry and academia to give insight in the future of implanted electrodes, implanted amplifiers and wireless transmission to actuators. This workshop is designed to bring together engineers, clinicians and neuroscientists. The audience is invited to share experiences with clinical implant studies. The discussed technologies are in different phases of readiness for clinical evaluation. Therefore, the focus of the workshop will shift from use tomorrow to use in a further future. Finally, the workshop organizer shares experience and will give guidelines to implement the latest technology in a clinical setting and into the homes of users.

Intended Audience

All BCI researchers having experience or with aspiring working with implanted BCIs, be it engineers, clinicians or neuroscientists. The experienced implant researchers share their know-how and will learn about new developments and avenues for future research, aspiring implant researchers will learn from examples of successful implant research and will gain knowledge how to implement new technologies.

Learning Objectives

1. Participants will be able to identify four essential ingredients of sensor technologies.2. Participants will be able to identify four essential ingredients of amplifier and transmission

technology3. Participants will be able to list 8 essential requirements for a clinical study

WS 18: Clinical Applications of Brain-Computer Interfaces in Neurorehabilitation

Presenters

Page 14: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

An H. Do, University of CaliforniaEvgeniy Kreydin, University of Southern CaliforniaCharles Liu, University of Southern CaliforniaZoran Nenadic, University of CaliforniaMarc Slutzky, Northwestern UniversityKarunseh Ganguly, University of CaliforniaSpencer Kellis, California Institute of Technology

Abstract

Brain-computer interfaces have increasingly been studied as a means to help those affected by neurological injuries (e.g. stroke, spinal cord injury [SCI], or traumatic brain injury [TBI]) to improve their functional outcome. In the clinical setting, BCI systems have attempted to restore motor functions that have been completely lost, or to help improve upon motor functions that are impaired, but still partially preserved. However, BCI-based rehabilitation is still very much an experimental approach. The first part of this workshop will explore the areas of clinical rehabilitation that are in urgent need of attention, examine how BCIs can potentially be used to address these areas, and discuss how BCIs are positioned in context of other emerging rehabilitation approaches. The second part of the workshop will focus on how close collaborations between clinicians, neuroscientists, and engineers can foster a meaningful approach to bridging the unmet needs in clinical rehabilitation.

Intended Audience

Clinicians (including physicians, physiotherapists), biomedical engineers, clinical-scientists, neuroscientists

Learning Objectives

1. Become familiar with the unmet clinical needs in neurorehabilitation2. Become familiar with how BCIs can be designed to address unmet needs and its context

amongst other existing or emerging treatment strategies3. Understand the importance of collaboration between clinicians, scientists, and engineers in

order to make meaningful progress in the field

SESSION 3- THURSDAY, MAY 24 (13:15-16:15)

WS 3: ECoG for control and mapping

Presenters

Christoph Guger, g.tec medical engineering GmbH, AustriaKyousuke Kamada, Asahikawa University, JapanMilena Korostenskaja, Florida Hospital for Children, USAJing Jin, East China University, ChinaKai Miller, Stanford University, USA

Abstract

Page 15: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

The Electrocorticogram provides beside alpha and beta oscillations also high resolution high-gamma activity. This high-gamma activation is very localized and has therefore a high spatial resolution beside the high temporal resolution. Therefore, the ECoG can be used for controlling prosthetic limbs, avatars or cursors, but can also be used to identify the eloquent cortex of a patient.

The workshop will show state-of-the art ECoG experiments for control and mapping and describe how the data acquisition, signal processing and experimental setup is done in the operating room and intensive care unit.

Intended Audience

- People interested in invasive ECoG based BCI technology- People interested in clinical studies with CCEPs and high-gamma mapping- People interested invasive avatar and prosthesis control

Learning Objectives

1. Participants will learn about current technology for ECoG BMI control2. Participants will be able to understand the target patient group3. Participants will learn about state-of-the art in ECoG based mapping and control

WS 5: Real-time BCI communication for non-verbal individuals with cerebral palsy: Challenges and Strategies for Progress

Presenters

Jane Huggins, PhD, University of MichiganJames A. Blackman, MD, MPH, Cerebral Palsy Alliance Research FoundationKatya Hill, PhD, CCC-SLP, University of PittsburghAdam Kirton, MD, University of CalgaryChristian Herff, PhD, University of Bremen

Abstract

Cerebral palsy (CP) is a neurodevelopmental disability resulting in disordered movement. Up to 50% of individuals with CP have significant communication impairment as well. Even though much progress has been made through development of computerized augmentative and alternative communication (AAC) devices, for someone with severe motor impairments, use of these remains exceedingly cumbersome and time-consuming. A thought-to-speech device would be life-changing, not only for persons with CP, but other neurological conditions and diseases (e.g. stroke, traumatic brain injury). However, most current efforts are directed at decoding intended speech from the motor cortex instead of from language areas of the brain. Thus, they assume the existence of typically developed motor areas of the brain devoted to producing speech. Unless specific attention is paid to the case of children born without the capacity for verbalization, the reliance on these assumptions could make a breakthrough in BCI for direct decoding of intended speech useless for people with CP. This workshop is intended to ensure that the special considerations necessary for thought-to-speech work for people with CP are considered early in the development process by creating a research strategy and roadmap for developing a real-time BCI for communication among individuals with CP.

Intended Audience

Page 16: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

Individuals interested in the challenges of developing a BCI for persons with Cerebral Palsy and other childhood-onset neurological disorders with severe speech/communication impairment.

Learning Objectives

1. Audience will be able to differentiate normal speech development in CP versus typical.2. Audience will identify three obstacles to developing a BCI for communication in CP.3. Audience will develop five new research questions to accelerate research in BCI for

individuals with communication impairment due to CP.

WS 10: Tools for establishing neuroadaptive technology through passive BCIs

Presenters

Dr. Thorsten O. Zander, Zander LaboratoriesDr. David Medine, Brain ProductsDr. Martijn Schreuder, ANTDr. Olva Krigolson, University of VictoriaLena M. Andreessen, Technische Universität BerlinLaurens R. Krol, Technische Universität Berlin

Abstract

This workshop is aimed at participants interested in applications of BCI technology in HumanComputer Interaction for users without disabilities. It is intended to represent the Society for Neuroadaptive Technology [1] and to support the communication between this society and BCI researchers.

Dr. Thorsten O. Zander (Team PhyPA) will present and discuss the aims of passive BCI research, Neuroadaptive Technology and modern Human-Computer Interaction [2]. This part represents a forum to identify synergies and develop new ideas. Dr. David Medine (Brain Products) will present technological developments for passive and hybrid BCIs by presenting the LabStreamingLayer (LSL, [3]), an open-source software project for synchronized, multi-modal data streaming and recording. Dr. Martijn Schreuder (ANT) and Dr. Medine will discuss the application of dry electrode systems in realistic scenarios and what hurdles need to be taken while recording EEG in the wild. Dr. Olav Krigolson (University of Victoria) will discuss the application of Neuroadaptive Technology in the field of Neuroeconomics and relate this specific area to the current state of the art in the BCI field.

The workshop will conclude with a live demonstration of M(eye)ndtris [4], a neuroadaptive version of Tetris which will be demonstrated with different EEG systems, including dry electrodes.

[1] www.neuroadaptive.org

[2] Zander, T. O., Krol, L. R., Birbaumer, N. P., & Gramann, K. (2016). Neuroadaptive technology enables implicit cursor control based on medial prefrontal cortex activity. Proceedings of the National Academy of Sciences (PNAS), 201605155.

[3] Kothe, C. (2014). Lab streaming layer (lsl). https://github. com/sccn/labstreaminglayer. Accessed on October, 26, 2015.

[4] Krol, L. R., Freytag S.-C., Zander, T. O., Meyendtris: A Hands-Free, Multimodal Tetris Clone using Eye Tracking and Passive BCI for Intuitive Neuroadaptive Gaming, In Proceedings of the 2017 ACM International Conference on Multimodal Interaction. ACM.

Page 17: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

Intended Audience

PhD students and senior researchers who are interested in applying BCI technology in a broad way, including users without disabilities.

Learning Objectives

1. Elaborate synergies between the fields of Brain-Computer Interface research, HumanComputer Interaction and Neuroeconomics.

2. Discuss concepts aiming at establishing Neuroadaptive Technologies in real-world scenarios.3. Learn about and apply tools for passive Brain-Computer Interfacing.

WS 12: Neurofeedback during Artistic Expression as Therapy

Presenters

Stephanie Scott, Colorado State UniversityChuck Anderson, Colorado State UniversityJuliet King, Indiana University School of MedicineLeslie Grace, MITRosa Mikeal Martey, Colorado State University

Abstract

This project explores the relationship between EEG and painting to determine whether neural oscillations can be modified as a result of the creative process. It investigates the existing relationships between art, science and technology when applied to the medium of painting, and examines what EEG band modifications can be implemented towards existing integrative efforts between neurofeedback technologies and art applications. Conceptually, these efforts are designed to promote collaboration, artistic self-awareness, rehabilitation and recovery, through the integration of a responsive closed neurofeedback loop during the process of painting. Other forms of artistic expression will be included, such as music, depending on the interests of workshop collaborators and attendees.

Intended Audience

This workshop is intended for all audiences who are interested in exploring new therapy applications for neurofeedback systems and tapping into less-explored avenues for obtaining insurance coverage for these devices, as well as those who are curious about how newer, more accurate noninvasive EEG sensing technologies can add capabilities to neurofeedback for therapy.

Possible audiences: BCI users, BCI researchers, Care Givers, Computer Scientists, Neurologists, Psychologists, Engineers, Communicators, Clinical Rehabilitation Specialists, Therapists and Art Therapists

Learning Objectives

1. Participants will be able to identify key benefits of engaging the arts and sciences to promote beneficial and therapeutic applications of BCI systems.

2. Participants will participate in, and gain insight from, discussions into how BCI and neurofeedback systems can extend a rehabilitative reach into different populations of users through less-explored therapy applications.

Page 18: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

3. Participants will be able to identify key characteristics concerning how the use of new EEG sensing technology will lead to more accurate BCI and neurofeedback data and research.

WS 16: Unsupervised Learning for BCIs

Presenters

Michael Tangermann, University of FreiburgDavid Hübner, University of FreiburgPieter-Jan Kindermans, GoogleInaki Iturrate

Abstract

A key challenge in BCIs is to tune a brain signal decoder to reliably detect a user’s state or intention. Traditional approaches achieve this by performing a time consuming supervised calibration session to train the decoder for each session or when changes in the neural signals are observed during a session. Recently, unsupervised learning methods for BCIs have been proposed that can update an existing decoder (e.g. initialized from transfer learning) or learn to decode the signals from scratch during the actual usage of a BCI.

The goals of the workshop are manifold. Different talks aim at providing an overview of state-of-the-art methods in unsupervised learning and their underlying principles. This should foster the combination of different unsupervised learning principles by machine learning researchers. It should also help BCI practitioners to identify scenarios in which unsupervised learning might be beneficial. Additionally, we want to identify reasons why unsupervised methods are rarely used by BCI practitioners and how the adoption can be accelerated.

Intended Audience

(A) The BCI practitioner, who can benefit from incorporating unsupervised learning in his BCI toolbox/workflow.

(B) The machine learning expert who develops new unsupervised learning methods.

Learning Objectives

1. Participants can identify scenarios in which unsupervised learning is applicable or not applicable.

2. Participant will be able to identify at least four different principles that allow for unsupervised learning.

3. Participants will partially be able to understand the pros and cons of the different unsupervised learning approaches.

WS 20: Perception of Sensation Restored through Neural Interfaces

Presenters

Jen Collinger, University of PittsburghDev Sarma, University of PittsburghJeff Ojemann, University of WashingtonDustin Tyler, Case Western Reserve University

Page 19: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

Luke Bashford, California Institute of TechnologyMichelle Salas, California Institute of TechnologyRob Gaunt, University of PittsburghTucker Tomlinson, Northwestern UniversitySliman Bensmaia, University of Chicago

Abstract

This workshop seeks to highlight current work related to the restoration of somatosensation through neural interfaces. Specifically, we will focus on how electrically-evoked sensations through peripheral and cortical neural interfaces are perceived by human subjects. What do they feel like, and why? The ultimate goal of restored somatosensation is to reproduce natural and useful feedback. We will discuss progress towards that goal and limitations with currently available technology. We will discuss how fundamental neuroscience and emerging technologies can be used to inform the development of biomimetic neuroprosthetics.

Intended Audience

Students, Trainees, Investigators, Clinicians

Learning Objectives

1. Participants will be able to describe different types of neural interfaces for restoring sensation

2. Participants will be able to describe the current state of restored sensation using neural interfaces

3. Participants will understand how biomimetic approaches may be used in the development of neural interfaces for restoring sensation

WS 21: From the lab into the wild: shaping methods and technologies for large-scale BCI research

Presenters

Matthias Hohmann, Max Planck Institute for Intelligent SystemsMoritz Grosse-Wentrup, Ludwig Maximilians Universität MünchenTim Mullen, IntheonConor Russomanno, Open BCIAJ Keller, Push the World LLCAlexandre Barachant, GrenobleVinay Jayaram, Max Planck Institute for Intelligent Systems

Abstract

Bringing brain-computer interfaces from the lab environment into the homes of patients and consumers has been challenging due to the lack of concrete tools to create user-friendly systems that allow for reproducible science and hypothesis testing. In this workshop, we will highlight novel technologies that may be key to enabling robust communication and control for patients, and breaking the limitations of current research in terms of sample size and longitudinal data.

The workshop will feature the following technologies:

- Cloud-based, real-time signal processing

Page 20: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

- Low-cost, wireless EEG hardware- Automated signal quality checking- MOABB: A benchmarking tool for systematic evaluation and ranking of BCI algorithms- An iOS prototype application for unsupervised, longitudinal EEG studies

During the group discussion, attendees will have the opportunity to identify potential features for future versions and evaluate them based on their relevance to the BCI community. The results will be discussed and incorporated into future releases. Our goal is to enable the community to conduct large-scale BCI studies by actively shaping the development process of the presented technologies together.

Intended Audience

The workshop is open to all audiences and a variety of backgrounds is actively encouraged to bring together different opinions on the presented technology and future developments.

Learning Objectives

1. Participants will learn about three new technologies that will help to translate their lab-BCIs into home-use systems

2. Participants will learn about automated signal quality checking for unsupervised BCI research3. Participants will learn about a new benchmark platform that will help ranking and selecting

classification algorithms4. Participants will get an understanding of the most urgent limitations when designing home

use BCIs during the group discussion5. Participants will actively shape the development of future technologies to suit their research

needs

WS 24: Standards for Neurotechnologies and Brain-Machine Interfacing

Presenters

Ricardo Chavarriaga, Ecole Polytechnique Fédérale de Lausanne Walter Besio, University of Rhode IslandCarole Carey, IEEE EMBCJosé Contreras-Vidal, University of Houston (to be confirmed)Tim Mullen, IntheonAnder Ramos-Murguialday, Tecnalia, Univertity of TübingenAbstract

The field of Brain-Machine Interfacing (BMI) is going through a very exciting period where the state-of-the-art in research is currently being tested on its intended end-users. This translation from research prototypes to viable clinical or consumer solutions entails multiple challenges. Furthermore, the possibility of deploying and commercializing BMI-based solutions requires researchers, manufacturers, and regulatory agencies to ensure these devices comply with well-defined criteria on their safety and effectiveness. In consequence, there is an increased interest on development of appropriate standards for BMI systems. Such endeavor is not trivial and requires all stakeholders to join efforts to identify priority areas that require standardization, and to devise incentives for adopting these standards early on at the development process. This workshop aims at discussing these topics and identify important elements to consider for proper standardization of both clinical and consumer applications. This workshop is technically sponsored by the IEEE Industry connections

Page 21: SESSION 1- WEDNESDAY, MAY 23 (9:00-12:00)bcisociety.org/.../04/BCI2018WorkshopSummaryWeb.docx · Web viewClosed-loop control of multi-pad FES electrodes allow ... the most common

group on neurotechnologies and the IEEE Brain Initiative. The workshop will be composed of invited presentations (15 min) and short spotlight presentations (5 min) from the attendees, complemented with group discussions. We intend to generate a clear picture of current challenges for standardization of neurotechnologies and possible strategies to overcome them.

Intended Audience

The workshop is aimed at all individuals interested in the translation of neurotechnologiesonto clinical and consumer applications. These include novice and experienced researchers, potential innovators and entrepreneurs, as well as representatives of the industry and regulatory bodies among others.

Learning Objectives

1. Participants will improve their knowledge about current situation and procedures regarding standards relevant to neurotechnologies and brain-machine interfaces

2. Participants will acknowledge the importance that standards have in the successful translation of research onto clinical and consumer applications. In particular, the need of taking these standards into account at early stages of development

3. Participants will be able to identify priority areas where standardization can help to expedite deployment of viable clinical and commercial neurotechnology-based products