Neural Interfaces Workshop 2005

September 7 – 9, 2005
Bethesda, MD

Over 500 attendees gathered at the Hyatt Regency Hotel in Bethesda, MD, to participate in the Neural Interfaces Workshop on September 7-9, 2005. As in 2004, the Workshop combined the 36th Annual Neural Prosthesis Workshop and the annual meeting of the National Institutes of Health's (NIH) Deep Brain Stimulation (DBS) Consortium.

Support for the Workshop came from the following Institutes within the NIH: the National Institute of Neurological Disorders and Stroke (NINDS), the Office of Rare Diseases (ORD), the National Institute on Aging (NIA), the National Institute of Biomedical Imaging and Bioengineering (NIBIB), the National Institute of Child Health and Human Development (NICHD), and the National Institute on Deafness and Other Communication Disorders (NIDCD).

The Workshop was organized around six plenary sessions: Progress in Deep Brain Stimulation, Novel Interface Technologies for Stimulation, Surgical Considerations for Neural Interfaces, Chronic Recording Microelectrodes, Neural Interfaces for Sensory Information, Spinal Cord Interfaces, and Future Efforts in Neural Interfaces. A highlight of the meeting was the dedicated poster sessions consisting of nearly 130 posters where valuable discussions and new collaborations were cultivated. This report is not intended to be completely comprehensive, but to highlight several discussions and a few presentations from the plenary sessions. The Workshop agenda and abstracts are available from the National Institute for Neurological Disorders and Stroke (NINDS) Neural Prosthesis Program web-site.

Dr. Story Landis, Director of the NINDS, welcomed the assembled audience of engineers, basic scientists, and clinicians. In her opening comments, she recognized the investigators who started this field for their vision of recording and decoding signals from the brain and translating the resulting findings to help patients with neurological disorders. Dr. Landis also commented on the more recent findings that DBS holds potential benefits for disorders beyond Parkinson's disease, such as depression. In opening comments, Mr. Jeffrey Martin shared his personal experience with DBS. His insights as a patient not only affirmed the benefits of neurotechnology, but also challenged the Workshop participants to continue to advance technologies and applications in DBS.

The goal of the first plenary session, entitled "Progress in Deep Brain Stimulation" and moderated by Dr. Eugene Oliver of NINDS, was to review the progress achieved within the DBS portfolio. Indeed, the investment in DBS research has spanned a diverse range of topics from how to assess and quantify efficacy of DBS in relieving movement dysfunction to the study of the psychological impact of DBS. While it is clear that significant progress has been made in recent years, a recurring point from many of the presenters was the need to gain a better understanding of the pathophysiology of movement disorders, which appear to be manifested in disruptions of normal sensorimotor network function. Two presentations focused on how to quantitatively characterize the therapeutic effects of DBS in rodents as well as in humans. Dr. Jing-Yu Chang, Wake Forest University, reported on his efforts to develop a rodent model of Parkinson's disease to explore the therapeutic mechanisms of DBS. He described the implementation of two tests for the rodent model: a treadmill motor task for forced movement, where DBS significantly improves both stance and swing dynamics, and an asymmetry test, where DBS improves spontaneous activity involving the lesion-affected limb. Dr. Daniel Corcos, University of Illinois at Chicago, presented his work concerning the effects of subthalamic nucleus (STN) stimulation on tremor, rigidity, and bradykinesia. His data quantitatively demonstrated that stimulation of the STN greatly improves the movement speed of both the elbow and ankle joint in patients with Parkinson's disease, approaching performance levels of healthy individuals. The basis of this improvement appears to be through increased activation of both agonist and antagonist muscles, as indicated by concurrent electromyographic measurements.

Dr. Dieter Jaeger, Emory University, addressed issues regarding the mechanisms of DBS action. His group used anesthetized rats where antidromic stimulation of the STN produced a dampening of the electroencephalographic potentials, consistent with the observed therapeutic effect of DBS. Dr. Robert Turner, University of California, San Francisco, argued that a comprehensive understanding of the effects of DBS requires detailed information simultaneous recordings using microelectrode arrays. By monitoring multiple locations simultaneously, he demonstrated that DBS alters pallidal somatic activity to abolish oscillatory and burst discharges characteristic of Parkinson's disease. Dr. Turner's observations suggest that STN DBS affects motor cortical activity through antidromic stimulation.

Initiation of movement, which is often internal and self-timed, is difficult for individuals suffering from Parkinson's disease. Dr. John Assad, Harvard Medical School, described his work investigating cortical and sub-cortical structures in movement initiation. Using multi-site recordings in behaving macaques, self-timed movements, but not reaction movements triggered through training, were found to be preceded by increased activity in the parietal cortex and sensorimotor putamen hundreds of milliseconds before movements were initiated. These results have implications for the design of neural prostheses, where systems using control signals derived from brain structures involved in self-timed movements may prove most effective.

Dr. Marjan Jahanashi, University of London, presented her work characterizing the effects of STN DBS on mood and cognition. Over all, STN DBS produces few adverse effects on cognition and some improvements on mood. One of the problems identified with DBS was the negative impact on verbal fluency, possibly due to the spreading of electrical stimulation beyond localized pathologic regions. During a brief platform presentation, Dr. Jeffrey Wertheimer, Wayne State University, presented survey results consistent with the positive effects of DBS, with the important caveat that DBS can negatively impact verbal fluency.

Dr. Jerrold Vitek, the Cleveland Clinic, discussed current clinical challenges for DBS in movement disorders. He highlighted numerous issues that may contribute to the inconsistency in the benefit derived from DBS. Among those issues that involve scientific design, it appears that the substantial methodological differences in many DBS studies often make interpretations complicated. Dr. Vitek also raised a series of current technological limitations including the inability for high resolution imaging following surgery due to electrode materials, the lack of telemetry for the implantable pulse generators (IPGs) and the need for extended battery life or rechargeable battery systems. His recommendations included expanding the capabilities of IPGs to enable the evaluation of novel stimulation waveforms, exploring the possibility of "smart stimulators" that have the capacity for dynamic internal adjustments, and well-controlled clinical trials for generation of class I evidence.

Several presentations during the plenary session entitled "Novel Interface Technologies for Stimulation," which was moderated by Dr. Joseph J. Pancrazio of NINDS, focused on the development of novel stimulation technologies that could impact both the DBS and neural prosthetics fields. Future DBS systems with greater precision of current delivery may reduce side-effects, such as the negative effects on speech fluency. Dr. Jit Muthuswamy, Arizona State University, is developing DBS microelectrode technology coupled to micro-actuators to enable precise and robust electrode placement. With the actuators and electrodes embedded in the same architecture, it is anticipated that the time required for surgery could be significantly reduced and it would be possible to easily adjust insertion depth post surgery. Dr. Jun Li, NASA Ames Research Center, presented early work on the development of a nanoelectrode array for assessing cellular physiology. The chip under development utilizes aligned carbon nanofibers to perform electrochemical recording through amperometry, as well as cell stimulation and recording. The ability to deliver more precise and complex stimulation patterns may be enhanced through the use of high density arrays. Mr. Scott Corbett, Advanced Cochlear Systems, presented work on the implementation of a next generation cochlear prosthetic stimulation array based on the use of liquid crystal polymers (LCPs) as a dielectric material. The advantages of using LCPs include ease of manufacturability through injection molding and biocompatibility. Furthermore, the weakest link for current high density arrays is the packaging/interconnect interface between wires and thin-film lithography-based arrays; using LCPs may result in a more durable polymer-based planar circuit technology.

A session entitled "Future Opportunities for DBS" involved a panel lead by Dr. Warren Gill, Duke University, and included Dr. Steven Shapiro, Virginia Commonwealth University, and Dr. Vitek. Dr. Shapiro began the session by discussing the application of DBS to a disorder of newborn infants, kernicterus, which is produced by excessive jaundice and manifested as a static, secondary dystonia associated with deafness. The main point of the presentation was that potential future applications for DBS may include the pediatric population, for which the technological requirements for long term implanted systems may be different than those established for Parkinson's disease, a disorder associated with adult and aged populations. In addition, the combination of DBS and cochlear prosthetics for children suffering from the effects of kernicterus raises the question of how to design and manage multiple neural interface technologies within a single patient for long periods of time.

The second day of the workshop began with an important discussion involving Mr. Laszlo Nagy, a high level quadriplegic with an implanted respiratory pacemaker that has freed him from use of a ventilator. Mr. Nagy communicated key patient concerns, including the invasiveness of some surgical solutions, saying he would choose to use speech recognition software over an implanted microchip to operate a computer. In response to the question of what capability he would require before he would be willing to pursue an implanted microelectrode array technology such as a brain machine interface, Mr. Nagy indicated that he would insist on restoration of a substantial function, such as the capability to use his own limb to feed himself.

Dr. Michael Weinrich of the NICHD moderated the session entitled "Surgical Considerations for Neural Interfaces," during which two neurosurgeons integrally involved in the application of neural prostheses addressed surgical considerations for neural interfaces. Dr. Gerhard Friehs, Brown University, described the on-going pilot human trial with the Cyberkinetics BrainGate system. Cyberkinetics, which has completed one year of study with their first patient subject, described a plan to eventually develop a wireless implementation of the BrainGate system. Dr. Michael Keith, Case Western Reserve University (CWRU), discussed surgical considerations for FNS systems and conveyed that the capability provided by FNS is only one of several factors in the decision of a paralyzed individual to opt for the technology, including the long time course for surgery and rehabilitation. Progress at CWRU regarding minimization of the number of wire leads, reduction in size, and improved power efficiency were outlined. Dr. Keith offered the vision of a neural interfaces "intranet" where multiple implanted devices communicate and share resources, such as power, to address the deficits of multiple physiologic systems affected during paralysis.

Progress reports were offered by Dr. Kensall Wise, University of Michigan, and Dr. Florian Solzbacher, University of Utah, during the plenary session entitled "Chronic Recording Microelectrodes" and moderated by Dr. Pancrazio. Both groups are working on different chronic electrode recording systems under contract with NINDS. Their goal is to develop microelectrode arrays that will be capable of robust recordings for periods as long as 6 months, to be demonstrated in non-human primates. Both groups have identified the external tethering of the implanted microelectrode arrays as a critical issue impeding long-term functionality. Therefore, both groups are developing microelectrode array systems that incorporate on-board amplification, spike detection, as well as wireless transmission of both power and data. The first performance phase of these two contractual efforts will be completed in Spring 2006, when the performers will be required to demonstrate recording capability of the wireless systems for at least two weeks in the non-human primate model.

Presentations during the plenary session entitled "Neural Interfaces for Sensory Information", which was moderated by Dr. Roger Miller of NIDCD, summarized ongoing work for three neural prostheses. The cochlear implant (CI) has provided a means for treating both adults and children that are severely hard-of-hearing or deaf. This device works by bypassing damaged cells within the inner ear and directly stimulating the surviving portions of the auditory nerve. As the number of CI users worldwide approaches 100,000, design elements from this neural prosthesis are being adapted to treat both visual and balance dysfunction. Presentations were given to summarize ongoing work for three neural prostheses at very different stages of development. Dr. Patricia Leake, University of California, San Francisco, offered insights on the use of cochlear implants taken from studies performed with animal models. Neurophysiology data taken from the inferior colliculus were reviewed to illustrate the influence of the duration of deafness and the stage of development have on neural afferents from the implanted cochlea. Dr. Leake also presented anatomical data supporting a trophic role arising from electrical stimulation that promotes survival of auditory neurons. Drs. Mark Humayan and James Weiland, University of California, Los Angeles, described progress from clinical studies with on an intraocular retinal prosthesis. An investigational device exemption was granted by the FDA for their study protocol, and videotaped demonstrations of patient tests were shown. The present device utilizes 16 platinum electrodes and provides cues that allow patients to discriminate visual object shape and motion with accuracies up to 80%. Dr. Charles Della Santina, Johns Hopkins University, presented his work on electrical stimulation of the vestibular nerve. Using an animal model of vestibular deficiency, he showed that stimulation with frequency modulated bipolar pulses delivered through an electrode placed in a single horizontal canal could induce compensatory vestibular-ocular reflex movement in the horizontal plane. Control over the spread of these stimulating currents into other branches of the vestibular nerve is an area of ongoing refinement. Future plans for a multichannel prototype that might be mounted on the head to encode three axes of rotation through electrical stimulation of the vestibular nerve were presented.

Dr. Naomi Kleitman of NINDS led the "Spinal Cord Interfaces" plenary session, which focused on technology and application of neural interfaces for the spinal cord. Dr. Mesut Sahin, New Jersey Institute of Technology, described a novel approach to interfacing with neural tissue, such as the spinal cord, which eliminates mechanical tethering associated with interconnects. His group has developed a photodiode based stimulator that converts incident near infrared wavelength light to an electrical stimulus. Preliminary data using a rat sciatic nerve model suggest that the stimulator may operate at depths of 1.5 mm from the laser source. Dr. Vivian Mushahwar, University of Alberta, presented exciting work demonstrating the utility of intraspinal microstimulation via implanted microwires to enable functional movements of hind limbs of adult cats with chronic spinal cord injury. In contrast with peripheral nerve stimulation, coordinated intraspinal microstimulation of motor neuron cell bodies in the ventral horn produced fatigue-resistant stepping movements, suggesting an advantage of intraspinal microstimulation over cuff-based peripheral nerve stimulation to restore limb movement. However, work in her laboratory and others indicated that the relative success of intraspinal stimulation for restoration of locomotion may not translate easily to other systems, such as micturition.

Two concurrent short courses were also held at the end of the second day. Ms. Marian Emr of NINDS held a session entitled "Speaking to the Press about Neural Interfaces." The second session entitled "Support Opportunities for Graduate Students and Post-Docs" was lead by Dr. Meredith Temple-O'Connor. After a presentation that introduced the NIH system to students and new investigators, Dr. Temple-O'Connor lead an informative question and answer period. She stressed the importance of contacting program staff before submission and during the grant process.

The "Future Efforts in Neural Interfaces" plenary session, moderated by Dr. Grace Peng of NIBIB, consisted of multiple forward-looking presentations that considered the potential impact of new emerging technologies on the future of neural interfaces research. Dr. Simon Giszter, Drexel University, working in collaboration with Dr. Frank Ko, described a newly developed braiding and weaving system capable of weaving micron dimension wires and nanofibers, producing a wide range of geometries and mechanical properties. With incorporation of electrically conductive substrates into the weave, novel electrode probe designs may be implemented.

Dr. Miguel Nicolelis, Duke University, presented three novel paradigms to explore the future of neural interfaces research. In the first paradigm, Dr. Nicolelis has applied a multi-electrode recording array approach to a dopamine transporter knockout mouse model which exhibits reversible Parkinsonian characteristics via blockade of dopamine synthesis. The second paradigm presented by Dr. Nicolelis involved studies in which implanted monkeys were provided with vibrotactile input as haptic feedback. He also presented a third paradigm related to development of bipedal locomotion using an animal model. In summary, these paradigms illustrate the utility of the multi-neuronal recording approach to address basic and translational questions of the central control of motor function in health and disease states.

Dr. Theodore Berger, University of Southern California, presented his work in the development of biomimetic electronic devices for cognitive function. This work is centered on of the development of novel mathematical models that encode the nonlinear dynamics of hippocampal neuronal networks. By integrating the modeling approach with electrode array recordings from hippocampal slices, Dr. Berger's group was able to demonstrate replication of some of the hippocampal functions by substituting the CA3 region of the hippocampus with a microchip implementation of the predictive mathematical models and is beginning to translate these methods to ensemble encoding studies in the whole brain, developing multi-input mathematical models that incorporate neuronal dynamics from pairs of synaptically connected cells. At a minimum, this biomimetic neural engineering highlights the strong potential that computational methods and models may have on enhancing the development of neural interface systems.

Two presentations offered alternative ways of stimulating neural tissues. Dr. Duco Jansen, Vanderbilt University, showed preliminary data indicating that it is possible to activate neural structure with low-level, pulsed infrared laser light. This system has the ability of applying a wide wavelength range for depth penetration in tissue to target nerve fasicles. The underlying mechanism of this effect is presently unclear, although it may involve photothermal effects evidenced by localized elevations in temperature. Dr. David Pepperberg, University of Illinois at Chicago, described efforts to construct neurotransmitter-mimicking molecular structures that tether and control the dynamics of the neurotransmitter ?-amino butyric acid (GABA) for neural prosthetic applications. Through tethering with azobenzene molecules, GABA could interact with postsynaptic receptors to elicit channel opening. Dr. Pepperberg's group is exploring the potential utility of this approach to create synthetic photosensors to stimulate inner retinal neurons via the GABA type C receptor.

Dr. Bruce Wheeler, University of Illinois at Urbana, reviewed advances in neural interfaces from studies using in vitro models. His group and others have developed reproducible patterns of neurons on two-dimensional architectures bearing microelectrode contacts for stimulation and recording. Dr. Wheeler emphasized the value of in vitro approaches to address fundamental and applied aspects of neuro-electronic interfaces and pointed out that not all in vitro models are equal; there are significant limitations in the use of transformed cell lines that typically fail to exhibit robust synaptic connectivity. Lastly, Dr. Wheeler also discussed the development of three-dimensional in vitro culture models through which the space around the electrode can be engineered and further studied.

The final session entitled "Future Directions for Neural Prosthesis" involved a panel led by Dr. Warren Grill, Duke University, including Dr. Berger, Dr. Della Santina, and Mr. Geoffrey Thrope, NDI Medical. Several goals and challenges were identified for future consideration relative to neural prosthetics. A major research and development goal is the implementation of a neuromotor prosthesis that would enable a paralyzed individual to control the movement of their own limb through volition. Achieving this goal would involve the combination of next generation FNS or intraspinal stimulation technologies coupled with robust and reliable brain machine interfaces that extract volitional signals. As implied by Mr. Nagy, metrics of success must include activities of daily living. It was noted during the panel discussion that a critical technology gap exists in the delivery of sensation to paralyzed individuals. In addition to returning the perception of sensation, sensory feedback would be anticipated to provide a performance benefit for neural prostheses for upper limb control.

In addition, Mr. Thrope offered perspectives from the private sector on how to achieve commercial success for neural prosthetics. His insights were derived from his previous marketing experience of the NeuroControl Freehand system, a surgically implanted device designed to restore hand function in people with quadriplegia by neuromuscular stimulation of forearm and hand muscles. In short, while end users of the Freehand technology considered the system a success and reimbursement from insurance companies was generally accepted, the sales volume did not meet the expectations of the financial backers. Mr. Thrope's plans for the next generation of the product, Freehand II, will address these business issues and surgical considerations of the patients.

Students were encouraged to attend the Neural Interfaces Workshop and apply for the Student Travel Assistance Program. Students selected for the program were recognized on the second day by Dr. Temple-O'Connor and Ms. Stephanie Fertig of NINDS. Of over 60 applications, 28 students were selected to receive assistance and most of these students presented a poster at the meeting. Overall student attendance to the Workshop was relatively high; approximately 24% of the participants identified themselves as students.

Planning is underway for the next Neural Interfaces Workshop in fall of 2006. Building on the success of the 2005 Workshop, the 2006 Workshop will continue to bring the DBS and Neural Prosthesis communities together. As plans are finalized over the next several months, updated information will be made available on the Neural Prosthesis Program web-site and Neural Prosthesis Program Listserv.