In the Spotlight Neuroengineering-ujt.pdf

IEEE REVIEWS IN BIOMEDICAL ENGINEERING, VOL. 2, 2009

In the Spotlight: Neuroengineering

Nitish Thakor , Fellow, IEEE

rapidly, expanding into new scientiﬁc frontiers as inno-

vative technologies are developed for research and for clinical

applications. The “hot” areas, at least as measured by popularity

and visibility, continue to be the ﬁelds of brain–computer inter-

face (BCI) or brain–machine interface (BMI), the application of

BMI to neural prosthesis, deep brain stimulation (DBS), neural

interface technologies, and brain imaging. While this list is cer-

tainly not exclusive, the choice of speakers and the subject of

their talks explored at major national and international confer-

ences evidence popularity of these topics. This article reviews

some of the highlights of what has been covered and explored

at various conferences and published in major journals.

IEEE Engineering in Medicine and Biology, the premier

conference in the ﬁeld of Biomedical Engineering, took place

in Minneapolis, MN, in August 2009, and featured a plenary

talk by Dr. Gary Glover from Stanford University on “MR

imaging of brain function: Challenges, opportunities and

questions.” Glover reviewed the state of the art technology

of functional magnetic resonance imaging (fMRI) as well as

exciting emerging applications in the social sciences as well

as biosciences. Among the emerging applications include the

ﬁelds of economics (neuromarketing), lie detection, psycho-

analysis, and biofeedback. At the same time, Glover illuminated

the ethical questions of misuse and exaggerated claims reached

using these techniques. In the Neuroengineering Track of the

conference, Dr. Andrew Schwartz from Pittsburgh University

gave a keynote talk on “Useful signals from motor cortex”

where he described how animals could control robots’ move-

ment with direct brain control in a self-feeding task. Schwartz

described his study in which monkeys controlled a robotic arm

continuously in 3-D space to reach out to food and retrieve it to

their mouths [1]. He summarized the research that increasingly

demonstrates that neurons provide a high ﬁdelity representa-

tion of the intended behavior of the limb. This basic work in

primates lays the foundation for providing natural movement in

prosthetic limbs. Dr. Jose Principe from University of Florida

gave a keynote talk on “Toward cognitive neuroprosthesis” that

reviewed his team’s recent work on a new “co-adaptive” close

loop paradigm for brain machine interfaces (BMIs) based on

reinforcement learning. The co-adaptive model of BMI utilizes

an environment wherein the computer algorithms as well as

the brain of the subject are interactive: the computer algorithm

would learn and adapt based on subject behavior, and the

subject, rodent brain in this case, would learn and adapt from

interactions with the environment.

The IEEE Special Topics Conference in Neural Engineering

took place in Turkey in May of 2009. The Plenary Speaker, Dr.

Jack Gallant, gave a talk on “Let’s see what you think! Bayesian

reconstruction of perceptual experiences from human brain

activity.” Gallant presented a new Bayesian decoding model

that can reconstruct natural images that were seen by an ob-

server from the brain activity measured using fMRI. This is an

exciting area of research going beyond the conventional BMI to

explore whether brain signals or images themselves represent or

describe the world “as the brain sees.” Reverse engineering the

fMRI may help clarify how distinct representations in different

parts of the brain can be combined to provide a coherent re-

construction of the visual world and open the window into how

brain processes visual perception. A second plenary speaker,

Dr. Arto Nurmikko, presented “Developments in implantable

wireless cortical interfaces for neural prosthetics.” The tech-

nology for neural interface and recording continues to progress

rapidly. Nurmikko’s talk selectively highlighted microscale

neural probes and wireless implantable active microchips. He

presented the challenge of building transcutaneous telemetry

systems for “broadcasting” multichannel neural signals. He

also presented the recent advances in recording and deciphering

Manuscript received October 02, 2009; revised October 05, 2009. Current

version published November 18, 2009.

N. Thakor is with Department of Biomedical Engineering, The Johns Hopkins

University, Baltimore, MD 21205 USA (e-mail: nthakor@jhu.edu).

Digital Object Identiﬁer 10.1109/RBME.2009.2034697

AUTHORIZED LICENSED USE LIMITED TO: IEEE XPLORE. DOWNLOADED ON MAY 13,2010 AT 11:47:36 UTC FROM IEEE XPLORE. RESTRICTIONS APPLY.

N EUROENGINEERING as a discipline continues to grow

IEEE REVIEWS IN BIOMEDICAL ENGINEERING, VOL. 2, 2009

selected command signals issued from the motor cortex by

small targeted population of neurons for speciﬁc motor tasks.

IEEE T RANSACTIONS ON N EURAL S YSTEMS AND

R EHABILITATION E NGINEERING , a ﬂagship journal of the IEEE

Engineering in Medicine and Biology Society, continues to

publish papers with high citation rates and impact. The paper

by K. R. Muller, C. W. Anderson, and G. E. Birch, “Linear

and nonlinear methods for brain-computer interfaces,” [2] was

selected as the Best Paper for receiving most citations in the

past ﬁve years. Another highly cited paper over the past ﬁve

years, “Comparison of linear, nonlinear, and feature selection

methods for EEG signal classiﬁcation,” by D. Garrett, D. A.

Peterson, C. W. Anderson, and M. H. Thaut [3] also gives an

indication of community interest in developing advanced EEG

signal processing methods, particularly for the application to

brain computer interface applications.

Interest in two sub-disciplines of the Neuroengineering ﬁeld,

BMI and DBS, remains high. This is reﬂected by topics of

the plenary symposia and workshops as well as a very high

attendance at the annual conference of IEEE Engineering in

Medicine and Biology Society. The workshop on “Beyond

brain machine interface: Motor, cognitive and virtual” orga-

nized by this author, was inaugurated by Dr. Elmar Schmeisser

from the Army Research Laboratory and included presentations

in four key areas: Neural Interface and Neural Prosthesis, Motor

Prosthetics and Rehabilitation, Cognitive Interface, and Virtual

Reality. The workshop was led by the author of this article with

an “Overview of neural interface technologies—Noninvasive

and invasive” and by Dr. Jiping He from University of Arizona

with an overview of “Neurophysiological Foundations.” This

workshop highlighted emerging trends, taking the ﬁeld of

Brain–Machine Interface beyond the connection to the com-

puter or controlling prostheses. Emerging work in the ﬁeld is

now exploring how BMI can be used to interact with physical

as well as virtual reality environments.

In addition to rigorous biomedical research on BCI and BMI,

innovators and entrepreneurs have also begun making an im-

pact on ﬁnding commercial applications. Toyota researchers in

Japan demonstrated an EEG-based BCI that allows the user to

control a wheelchair’s movements by thinking the commands

with a response time of merely 125 milliseconds [4]. Neurosky,

Inc, (San Jose, CA, USA) has announced a video game platform

that allows users to control elements in the game via an EEG

recording electrode that rests on their forehead. IEEE Spectrum

Magazine recently reported on companies like Emotiv working

toward video game applications for BCI [5]. However, that ar-

ticle also casts a distinctly skeptical glance at the means and

success of the method and its potential technical and business

success. It is clear that the ﬁeld of BCI/BMI is accompanied by a

great deal of press and hype. Accordingly, students, researchers,

and even investors need to tread cautiously.

The nascent ﬁeld of neural prosthesis reported signiﬁcant sci-

entiﬁc advancements. In a paper in Nature , Moritz et al. [6]

demonstrated for the ﬁrst time that cortical activity in monkeys

can be used to directly stimulate and control previously para-

lyzed muscles on the monkey’s body via functional electrical

stimulation. This restored goal-directed movements to a para-

lyzed arm by re-establishing the connection between neurons

in the motor cortex and neurons in the target muscle while by-

passing normal physiological nerve transmission routes. More-

over, motor neurons and cortical neurons previously unassoci-

ated with movement performed this task equally well. These

ﬁndings hold great potential for those with paralyzed limbs as

a result of spinal cord injury or peripheral nerve damage to

regain intuitive control over their limbs. John Donoghue [7]

reviewed the progress towards building the neural interfaces,

and the challenges faced by the technology and the brain inter-

face. Despite the technical hurdles and our incomplete knowl-

edge of BMI, human studies have shown promise. Neural spike

activity signals recorded by probes implanted in humans have

been successfully decoded to control computer cursor velocity

in tetraplegic patients [8].

Another popular area of research with clinically high impact

and considerable industrial technology development is the ﬁeld

of deep brain stimulation (DBS). DBS has powerful clinical ap-

plications, such as targeting stimulation to subthalamic nuclei

to suppress Parkinsonian tremors. The popularity of the ﬁeld

was evidenced by the well attended pre-conference Workshop

on DBS organized by Aviva Abosch (University of Minnesota

Medical School, USA), Gregory F. Molnar (Medtronic, Inc.,

USA), and Gregory A. Worrell (Mayo Clinic, USA). The work-

shop organizers noted that DBS has been approved by the U.S.

Food and Drug Administration for the treatment of Parkinson’s

disease, essential tremor, and dystonia. To date, over 30 000 de-

vices have been implanted worldwide in humans. The workshop

led of with an overview of DBS by a leading Neurosurgeon,

Dr. Ali Rezai, from the Cleveland Clinic Foundation. The cov-

ered the topics of Regulatory History of DBS & Development

of the Field, In-Depth Clinical Outcome of DBS for Movement

Disorders, In-Depth Clinical Outcome of DBS for Obsessive-

Compulsive Disorder, Safety considerations—Procedural and

Radiofrequency Heating, Basal Ganglia Physiology and DBS,

Theories of Mechanism of Action of DBS, Electrical stimula-

tion for the treatment of psychiatric disorders, Brain Stimula-

tion Devices for Epilepsy, Device Technology Innovation.

This year has seen great progress in deciphering mental con-

tent from brain activity. An article in Science by Formisano et

al. [9] describes how their group was able to use fMRI to look at

the auditory cortex and determine if the subject was listening to

something that they had heard before and if it was from a speaker

whom they recognized. Their ﬁndings have shed light on details

pertaining to speech perception and the computational proper-

ties of the neural populations in the auditory cortex. A sim-

ilar brain-decoding milestone has been achieved with regards

to the visual cortex. Past fMRI studies have decoded position,

orientation, and object category from neural activity in the vi-

sual cortex. In addition, Kay et al. [10] reported in Nature that

they have developed a decoding method that goes even farther

and allows for identiﬁcation of the speciﬁc image viewed by the

subject out of a large set of novel natural images. This method is

based on analyzing the tuning of individual voxels for space, ori-

entation, and spatial frequency estimated from brain responses

evoked by natural images.

As it occasionally happens in science, unexpected discoveries

or advances in one ﬁeld suddenly take hold in another discipline.

Technology developed by basic scientists rapidly disseminates

and crosses over to novel applications. In a series of papers, Dr.

Karl Deisseroth’s group at Stanford and Dr. Edward Boyden’s

AUTHORIZED LICENSED USE LIMITED TO: IEEE XPLORE. DOWNLOADED ON MAY 13,2010 AT 11:47:36 UTC FROM IEEE XPLORE. RESTRICTIONS APPLY.

IEEE REVIEWS IN BIOMEDICAL ENGINEERING, VOL. 2, 2009

group at MIT have described molecular probes capable of using

light to activate neurons at high speeds and resolutions and ex-

plored some remarkable properties and novel applications.

In a paper published in Nature Neuroscience , “Bi-stable

neural state switches,” Berndt et al. [11] from Deisseroth’s

group describe optically switchable molecules called chan-

nelrhodopsins. Channelrhodopsins are a class of naturally

occurring as well as genetically engineered molecules. These

molecules are incorporated in cell membranes of cells, in-

cluding neurons, and using short light pulses can be turned

on or off with a millisecond-level temporal resolution. This

remarkable property can then be used to turn neurons on or off

with a high temporal resolution using light, previously a domain

of electrical excitation only. Since light can be delivered at a

distance (through laser light sources), rapidly (through scan-

ning), or even in vivo (using ﬁber optics) this molecular probe

has opened up exciting new capabilities for basic research a

broad range of in vivo applications.

In a paper published in Neuron , “Millisecond-timescale

optical control of neural dynamics in the nonhuman primate

brain,” Graybiel et al. [13] from Boyden’s group used lentivirus

to target such channelrhodopsins to target excitatory neurons of

the macaque cortex. They used ﬁber optics to couple the light

to the neurons in the cortex and speciﬁcally activated excita-

tory neurons in the neocortical neural networks. Remarkably

the molecules are expressed, and optically modulated, over

extended period of time (months so far), which opens up the

possibility of probing and modulating cortical neurons in vivo .

The authors speculate that this technique will advance appli-

cations to research on nonhuman primate neural computation,

cognition, and behavior, ultraprecise neurological and psychi-

atric therapeutics and even optical neural control of prosthetics.

In a paper published in Science , “Optical deconstruction of

Parkinsonian neural circuitry,” Gradinaru et al. [12] from Deis-

seroth’s group showed that the same optical technique could be

used to study the mechanism of DBS. The mechanism behind

the action of stimulation on neuronal circuits in these deep re-

gions is not fully known. Using solid-state optics, the authors

carried out studies in freely moving rodents exhibiting Parkinso-

nian symptoms to systematically drive or inhibit speciﬁc circuit

elements and found that therapeutic effects could be accounted

for by direct selective stimulation of afferent axons projecting to

thalamic nuclei. Indeed this opens the doors to study other dis-

ease circuits by selectively controlling neurons and neural net-

works in vivo in other disease models.

Research on the brain has a human element to it as well. Every

investigative procedure or therapy has its volunteers and pio-

neers. For example, Jesse Sullivan volunteered extensively to

demonstrate the function and success of peripheral nerve rein-

nervation surgery and to rewire the nerve from the amputee’s

residual limb to his chest. The ’“rewired” signals from the nerve

actuate the muscles on the chest and they in turn are decoded to

control an advanced dexterous prosthetic limb. Jonathon Kun-

niholm, also an amputee veteran of the recent wars, has very ar-

ticulately and passionately voiced the need for technological ad-

vancements and affordable and accessible solutions for the am-

putees [14]. On a sad note, a pioneer research volunteer, know as

“H. M.”, who helped revolutionize our knowledge about brain

and memory died last year. H. M. underwent an experimental

brain operation to correct a seizure disorder in 1953, but ended

up with an irreparable brain damage from the surgery. He de-

veloped profound amnesia, losing the ability to form new mem-

ories. Essentially, he could not form long term memories from

short term events and experiences. Experiments on behavioral

and brain imaging studies he volunteered for shed considerable

light on the role of the hippocampus in memory formation and

retention. This article therefore salutes the volunteers like Jesse,

Jonathon, and H.M. who are helping pave the frontiers of Neuro-

engineering research and development, including fundamental

understanding about the brain and its diseases and the techno-

logical innovations and solutions for therapy.

A CKNOWLEDGMENT

The author thanks N. Davidovics, N. Li, and Y. Choi for their

help with the literature and news review in preparation of this

article.

R EFERENCES

[1] M. Velliste, S. Perel, M. C. Spalding, A. S. Whitford, and A. B.

Schwartz, “Cortical control of a prosthetic arm for self-feeding,”

Nature , vol. 453, pp. 1098–1101, Jun. 19, 2008.

[2] K.-R. Muller, C. W. Anderson, and G. E. Birch, “Linear and nonlinear

methods for brain-computer interfaces,” IEEE Trans. Neural Systems

Rehab. Eng. , vol. 11, no. 2, pp. 165–169, Jun. 2003.

[3] D. Garrett, D. A. Peterson, C. W. Anderson, and M. H. Thaut, “Com-

parison of linear, nonlinear, and feature selection methods for EEG

signal classiﬁcation,” IEEE Trans. Neural Systems Rehab. Eng. , vol.

11, no. 2, pp. 141–144, Jun. 2003.

[4] [Online]. Available: http://electric-wheelchair-on.net/access-

ability-devices/toyota-demos-mind-controlled-electric-wheelchair/

[5] D. Heingartner, “Mental block: Emotiv says its game controller works

at the speed of thought, but it doesn’t,” IEEE Spectrum , vol. 46, pp.

42–43, Jan. 2009.

[6] C. T. Moritz, S. I. Perlmutter, and E. E. Fetz, “Direct control of paral-

ysed muscles by cortical neurons,” Nature , vol. 456, pp. 639–642, Dec.

4, 2008.

[7] J. P. Donoghue, “Bridging the brain to the world: A perspective on

neural interface systems,” Neuron , vol. 60, no. 3, pp. 511–521, Nov. 6,

2008.

[8] S.-P. Kim, J. D. Simeral, L. R. Hochberg, J. P. Donoghue, and M. J.

Black, “Neural control of computer cursor velocity by decoding motor

cortical spiking activity in humans with tetraplegia,” J. Neural Eng. ,

vol. 5, pp. 455–476, Nov. 17, 2008.

[9] E. Formisano, F. De Martino, M. Bonte, and R. Goebel, ““Who” is

saying “What”? Brain-based decoding of human voice and speech,”

Science , vol. 322, no. 5903, pp. 970–973, Nov. 7, 2008.

[10] K. N. Kay, T. Naselaris, R. J. Prenger, and J. L. Gallant, “Identifying

natural images from human brain activity,” Nature , vol. 452, pp.

352–355, Mar. 20, 2008.

[11] A. Berndt, O. Yizhar, L. A. Gunaydin, P. Hegemann, and K. Deis-

seroth, “Bi-stable neural state switches,” Nature Neurosci. , vol. 12, pp.

229–234, 2008.

[12] V. Gradinaru, M. Mogri, K. R. Thompson, J. M. Henderson, and K.

Deisseroth, “Optical deconstruction of parkinsonian neural circuitry,”

Science , vol. 324, no. 5925, pp. 354–359, Apr. 17, 2009.

[13] M. Graybiel, R. Desimone, and E. S. Boyden, “Millisecond-timescale

optical control of neural dynamics in the nonhuman primate brain,”

Neuron , vol. 62, no. 2, pp. 191–198, Apr. 30, 2009.

[14] J. Kuniholm, “Open arms,” IEEE Spectrum , vol. 46, pp. 36–41, Mar.

2009.

AUTHORIZED LICENSED USE LIMITED TO: IEEE XPLORE. DOWNLOADED ON MAY 13,2010 AT 11:47:36 UTC FROM IEEE XPLORE. RESTRICTIONS APPLY.

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: