Articles

Restoration of reaching and grasping movements through

brain-controlled muscle stimulation in a person with
tetraplegia: a proof-of-concept demonstration
A Bolu Ajiboye*, Francis R Willett*, Daniel R Young, William D Memberg, Brian A Murphy, Jonathan P Miller, Benjamin L Walter, Jennifer A Sweet,
Harry A Hoyen, Michael W Keith, P Hunter Peckham, John D Simeral, John P Donoghue, Leigh R Hochberg, Robert F Kirsch

Summary
Background People with chronic tetraplegia, due to high-cervical spinal cord injury, can regain limb movements Published Online
through coordinated electrical stimulation of peripheral muscles and nerves, known as functional electrical March 28, 2017
http://dx.doi.org/10.1016/
stimulation (FES). Users typically command FES systems through other preserved, but unrelated and limited in S0140-6736(17)30601-3
number, volitional movements (eg, facial muscle activity, head movements, shoulder shrugs). We report the findings
See Online/Comment
of an individual with traumatic high-cervical spinal cord injury who coordinated reaching and grasping movements http://dx.doi.org/10.1016/
using his own paralysed arm and hand, reanimated through implanted FES, and commanded using his own cortical S0140-6736(17)30562-7
signals through an intracortical braincomputer interface (iBCI). *Contributed equally
Department of Biomedical
Methods We recruited a participant into the BrainGate2 clinical trial, an ongoing study that obtains safety Engineering (A B Ajiboye PhD,
information regarding an intracortical neural interface device, and investigates the feasibility of people with F R Willett BS, D R Young BS,
W D Memberg MS,
tetraplegia controlling assistive devices using their cortical signals. Surgical procedures were performed at University B A Murphy PhD,
Hospitals Cleveland Medical Center (Cleveland, OH, USA). Study procedures and data analyses were performed at Prof P H Peckham PhD,
Case Western Reserve University (Cleveland, OH, USA) and the US Department of Veterans Affairs, Louis Stokes Prof R F Kirsch PhD) and School
Cleveland Veterans Affairs Medical Center (Cleveland, OH, USA). The study participant was a 53-year-old man with of Medicine (A B Ajiboye,
F R Willett, D R Young,
a spinal cord injury (cervical level 4, American Spinal Injury Association Impairment Scale category A). He received W D Memberg, B A Murphy,
two intracortical microelectrode arrays in the hand area of his motor cortex, and 4 months and 9 months later J P Miller MD, B L Walter MD,
received a total of 36 implanted percutaneous electrodes in his right upper and lower arm to electrically stimulate J A Sweet MD, H A Hoyen MD,
M W Keith MD,
his hand, elbow, and shoulder muscles. The participant used a motorised mobile arm support for gravitational
Prof P H Peckham,
assistance and to provide humeral abduction and adduction under cortical control. We assessed the participants Prof R F Kirsch), Case Western
ability to cortically command his paralysed arm to perform simple single-joint arm and hand movements and Reserve University, Cleveland,
functionally meaningful multi-joint movements. We compared iBCI control of his paralysed arm with that of a OH, USA; US Department of
Veterans Affairs, Louis Stokes
virtual three-dimensional arm. This study is registered with ClinicalTrials.gov, number NCT00912041.
Cleveland Veterans Affairs
Medical Center, Functional
Findings The intracortical implant occurred on Dec 1, 2014, and we are continuing to study the participant. The last Electrical Stimulation Center of
session included in this report was Nov 7, 2016. The point-to-point target acquisition sessions began on Oct 8, 2015 Excellence, Rehabilitation R&D
Service, Cleveland, OH, USA
(311 days after implant). The participant successfully cortically commanded single-joint and coordinated multi-joint
(A B Ajiboye, F R Willett,
arm movements for point-to-point target acquisitions (80100% accuracy), using first a virtual arm and second his D R Young, W D Memberg,
own arm animated by FES. Using his paralysed arm, the participant volitionally performed self-paced reaches to B A Murphy, J P Miller, B L Walter,
drink a mug of coffee (successfully completing 11 of 12 attempts within a single session 463 days after implant) and J A Sweet, Prof P H Peckham,
Prof R F Kirsch); Department of
feed himself (717 days after implant).
Neurology (B L Walter) and
Department of Neurological
Interpretation To our knowledge, this is the first report of a combined implanted FES+iBCI neuroprosthesis for Surgery (J P Miller, J A Sweet,
restoring both reaching and grasping movements to people with chronic tetraplegia due to spinal cord injury, and Prof R F Kirsch), University
Hospitals Cleveland Medical
represents a major advance, with a clear translational path, for clinically viable neuroprostheses for restoration of Center, Cleveland, OH, USA;
reaching and grasping after paralysis. Department of Orthopaedics,
MetroHealth Medical Center,
Funding National Institutes of Health, Department of Veterans Affairs. Cleveland, OH, USA (H A Hoyen,
M W Keith); School of
Engineering (J D Simeral PhD,
Introduction through skin surface, intramuscular, or nerve cuff Prof L R Hochberg MD), Brown
High-cervical spinal cord injury resulting in tetraplegia electrodes,13 and has successfully restored grasping to Institute for Brain Science
prevents affected individuals from performing reaching individuals with mid-level to low-level cervical spinal (J D Simeral,
Prof J P Donoghue PhD,
and grasping movements required for many activities of cord injury, who retained both volitional shoulder and L R Hochberg) and Department
daily living. Functional electrical stimulation (FES), in elbow movements to command stimulation.46 of Neuroscience
the absence of descending motor commands, applies Restoration of multi-joint reaching and grasping is (Prof J P Donoghue), Brown
spatiotemporal patterns of stimulation to peripheral more difficult in individuals with high-cervical spinal University, Providence, RI, USA;
Center for Neurorestoration
nerves and muscles to reanimate paralysed limbs for cord injury because the few available command and Neurotechnology,
restoration of lost functions. FES can be delivered options (sip-and-puff, eye tracking, retained head and

www.thelancet.com Published online March 28, 2017 http://dx.doi.org/10.1016/S0140-6736(17)30601-3 1

Descargado para Anonymous User (n/a) en ClinicalKey Espanol Colombia, Ecuador & Peru Flood Relief de ClinicalKey.es por Elsevier en abril 28, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
 Articles
Rehabilitation R&D Service,
Department of Veterans Affairs
Medical Center, Providence, RI,
USA (J D Simeral,
Prof J P Donoghue,
Prof L R Hochberg); Department
of Neurology, Massachusetts
General Hospital, Boston, MA,
USA (J D Simeral,
Prof L R Hochberg); Department
of Neurology, Harvard Medical
School, Boston, MA, USA
(L R Hochberg); and Wyss
Center for Bio and
Neuroengineering, Geneva,
Switzerland (Prof J P Donoghue)
Correspondence to:
Dr A Bolu Ajiboye, Department
of Biomedical Engineering, Case
Western Reserve University,
Cleveland, OH 44106, USA
aba20@case.edu
For more on the inclusion and
exclusion criteria see
https://clinicaltrials.gov/ct2/
show/NCT00912041?term=NCT
00912041&rank=1
2 www.thelancet.com Published online March 28, 2017 http://dx.doi.org/10.1016/S0140-6736(17)30601-3
Descargado para Anonymous User (n/a) en ClinicalKey Espanol Colombia, Ecuador & Peru Flood Relief de ClinicalKey.es por Elsevier en abril 28, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
Research in context
Evidence before this study
We initially searched PubMed using the search terms
(FES OR electrical stimulation) AND (BMI OR BCI OR to a person with chronic spinal cord injury that results in complete
brain-machine interface, OR brain-computer interface),
with no language or date restrictions. The date of our last
search was Nov 16, 2016. Our search resulted in a large
number of studies in people using predominantly
non-invasive braincomputer interfaces (BCIs) to command
non-focal surface stimulation to restore state-based, all-or-
nothing hand opening and closing. Other noteworthy studies
used non-invasive BCIs combined with an implanted
Freehand functional electrical stimulation (FES)
neuroprosthesis to again restore state-based, all-or-nothing
hand opening or closing. One study used an intracortical
microelectrode array with a surface FES system to restore
hand grasping alone to a person with mid-level cervical spinal
cord injury. Two non-human primate studies were of note
that showed restoration of continuous (graded) control of
implanted FES activation of wrist and hand function. Three
studies in individuals who were paralysed showed BCI control
of robotic arms. However, we found no studies that were
similar to this study, either in individuals with spinal cord
injury or non-human primate paralysis models, that restored
both continuous reaching and grasping functions via
electrical stimulation and also had a clear path to clinical
translation.
neck movements) are unintuitive, scale poorly for
commanding coordinated multi-joint movements, and
interfere with intact head and face function. Intracortical
braincomputer interfaces (iBCIs) that directly map
cortical activity to desired movement eschew the need
for retained volitional movement, thereby potentially
addressing these shortcomings. Intact non-human
primates79 and people with paralysis1013 have
successfully used iBCIs to command cursor movements
and reaching and grasping movements using robotic
limbs. Temporarily paralysed non-human primates
have used iBCIs to command implanted FES-actuated
wrist and grasping movements.14,15 A study published in
2016 used an iBCI coupled with surface electrical
stimulation to provide assistive hand grasping to an
individual with a C5/C6 spinal cord injury16 who
retained volitional shoulder and elbow function.
However, the 25-year-old Freehand implanted FES
system (NeuroControl Corporation, Valley View, OH)46
has already successfully restored hand grasping to
individuals who retained volitional arm function,
without requiring an iBCI. We report the findings of an
individual with chronic tetraplegia who used an
implanted FES system to make both reaching and
grasping movements, intuitively and effectively
commanded by an iBCI, with a translational path for
future clinical viability.
Added value of this study
Our study is the first to restore both reaching and grasping via FES
loss of arm and hand function. By using both an intracortical BCI
and percutaneous FES electrodes for muscle activation, as well as
a mobile arm support for gravitational assistance, we have shown
a proof-of-concept combined technology that allows users to
perform functional tasks that require coordinated reaching and
grasping. Although other non-invasive BCI and FES hand-only
systems have been proposed, none have been shown to be readily
adoptable for day-to-day use, and certainly not for restoring both
reaching and grasping. The present work has a clear path to
clinical translation because of the fully implantable FES
technology that already exists, and the continued efforts to
develop fully implanted and wireless BCI systems.
Implications of all the available evidence
Our results show the potential of combining implanted FES and
iBCI (with a mobile arm support) for restoring self-initiated
reaching and grasping movements to individuals with spinal
cord injuries that result in chronic paralysis. This work was a
crucial step for demonstrating feasibility. Future developments
of fully implanted systems, as well as developments in
advanced decoders and stimulators, might lead to enhanced
neuroprosthetic functional performance and greater
independence for individuals with paralysis.
Methods
Participant
We enrolled the participant (ID number T8) into the
BrainGate2 clinical trial (ClinicalTrials.gov, number
NCT00912041). Inclusion and exclusion criteria are avail
able online. Surgical procedures were performed at
University Hospitals Cleveland Medical Center
(Cleveland, OH, USA). Study procedures and data
analyses were performed at Case Western Reserve
University (Cleveland, OH, USA) and the US Department
of Veterans Affairs, Louis Stokes Cleveland Veterans
Affairs Medical Center (Cleveland, OH, USA). The
participant both consented verbally and gave written
informed consent through his power of attorney for
study procedures as approved by the Institutional Review
Boards of University Hospitals Cleveland Medical Center
(Cleveland, OH, USA) and Massachusetts General
Hospital (Boston, MA, USA).
Procedures
At the time of the intracortical implant (on Dec 1, 2014),
the participant was a 53-year-old man who had
experienced traumatic high-cervical spinal cord injury
(cervical level 4, American Spinal Injury Association
Impairment Scale category A) 8 years before enrolment.
On his dominant right side (contralateral to the
intracortical implant), he retained restricted and non-
 Articles

functional voluntary shoulder girdle motion, but no A Patient cable

voluntary glenohumeral, elbow, or hand function, and no
sensation below the shoulder. An implanted baclofen Neural activity

pump controlled spasticity of his dominant arm. Recording array

Two 96-channel microelectrode arrays (Blackrock
Microsystems, Salt Lake City, UT, USA)17 were implanted
into the hand area on the precentral gyrus18 of his motor
cortex (appendix p 3). After 4 months of using the iBCI to
command movements of a three-dimensional virtual Decoder
arm, the participant received, during two procedures
(125 days and 280 days after implant), 36 percutaneous
muscle-stimulating electrodes (Synapse Biomedical,
iBCI Implanted lead
Oberlin, OH, USA)19 in his right upper and lower arm,
FES
including four percutaneous anodic current return Electrode
electrodes, to restore finger and thumb (for a lateral hand External
grasp),20 wrist, elbow, and shoulder movements (see stimulator

appendix p 15 for a list of all implanted muscles). Starting Percutaneous

on day 142 after implant, all implanted muscles were lead connector
exercised with cyclical electrical stimulation patterns to
improve strength, range of motion, and fatigue
resistance. Exercise occurred 18 out of 45 weeks, Instrumented
goniometer
averaging 8 h per week spread over 23 days.
Figure 1A illustrates the FES+iBCI system. The iBCI
consisted of the implanted recording microelectrode Mobile arm support
arrays, with a neural decoder that translated recorded
cortical activity into command signals for controlling
muscle stimulation to produce coordinated reaching and
B
grasping movements. The FES system consisted of an Elbow stimulation pattern Wrist stimulation pattern
external stimulator that delivered charge-balanced, 200 Extension Flexion Extension Flexion
Electrodes Electrodes
biphasic, constant-current stimulation through Triceps1, Triceps2 ECRL, ECU, EDC1, EDC2
150 Biceps1, Biceps2
percutaneous electrodes to produce muscle contractions FCU
Pulse widths (s)

Biceps3
and subsequent limb movement. The stimulation had a
100
fixed current amplitude (20 mA) and frequency
(125 Hz), and a variable pulse duration of 0200 s. The
50
current pulse duration (known as pulse-width) applied
at a given electrode determined the strength of the
0
muscle contraction. The participant used a mobile 0 20 40 60 80 100 0 20 40 60 80 100
arm support (Focal Meditech, Tilburg, Netherlands) for Stimulation pattern (%) Stimulation pattern (%)

support against gravity and motorised humeral Hand stimulation pattern

abduction and adduction (also under cortical command), 200 Extension Flexion
because neither his residual shoulder motion nor deltoid Electrodes
FDSI EPL2
stimulation provided adequate humeral abduction and 150 ADP1, ADP2 EDC1, EDC2
Pulse widths (s)

adduction. Instrumented goniometers (Biometrics Ltd, EPL1

Ladysmith, VA, USA) on his elbow, wrist, and hand 100
measured joint motions, and an orientation sensor
quantified mobile arm support movements. 50
Neural decoders were calibrated daily at the beginning
of each experimental session to translate cortical activity 0
0 20 40 60 80 100
patterns into movement commands for a virtual reality Stimulation pattern (%)
arm or the FES-actuated arm. Daily recalibration helped
Figure 1: Overview of the FES+iBCI system
to account for day-to-day variability in the recorded
(A) Neural activity was recorded from two microelectrode arrays implanted in the motor cortex, which was then
cortical activity.21 The decoders used two neural features decoded into command signals that determined the amount of stimulation applied by the FES system and controlled
from each electrode of the intracortical arrays: unsorted the elevation of the mobile arm support. (B) Example stimulation patterns for the elbow, wrist, and hand. Stimulation
threshold crossing rates, determined by counting of all patterns convert the decoded command signals into the appropriate pulse widths to apply to each individual FES
electrode, which enables the participant to coordinate the action of multiple electrodes and muscles using only a
action potentials in a 20 ms time window that crossed a
single command. FES=functional electrical stimulation. iBCI=intracortical braincomputer interface. ECRL=extensor
preset noise threshold, and average spectral high carpi radialis longus. ECU=extensor carpi ulnaris. EDC=extensor digitorum communis. FCU=flexor carpi ulnaris.
frequency power (2503000 Hz) in a 20 ms time FDSI=flexor digitorum superficialis (index). ADP=adductor pollicis brevis. EPL=extensor pollicis longus.

www.thelancet.com Published online March 28, 2017 http://dx.doi.org/10.1016/S0140-6736(17)30601-3 3

Descargado para Anonymous User (n/a) en ClinicalKey Espanol Colombia, Ecuador & Peru Flood Relief de ClinicalKey.es por Elsevier en abril 28, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
 Articles

See Online for appendix window. The decoders used a linear transformation Comparison of cortical control of a virtual reality arm
function, similar to the Kalman filter used in previous versus an FES-actuated arm
iBCI applications,22 to map the neural features to three Figure 2A illustrates the timeline of surgeries and
movement commands. For the virtual arm, decoded sessions. As part of the BrainGate2 clinical trial, the
commands determined the instantaneous movement study participant performed various virtual reality arm
velocities for the virtual arm joints (shoulder, elbow, control sessions that were not directly related to the
wrist, grasp). For the FES arm, decoded commands present report. For this report, the participant performed
determined the change in the percentage activation of two types of sessions: sessions in which he controlled
stimulation patterns associated with elbow, wrist, or both the virtual reality arm and the FES system to
hand movements (or determined the actuation of the compare the performance between the two, and sessions
mobile arm support). The stimulation patterns made it in which he controlled the FES system to complete
easy for the participant to coordinate the activity of meaningful functional tasks. Data reported in this study
multiple electrodes in a graded fashion, with only a few are from 26 virtual reality versus FES comparison
command signals20 (appendix pp 1, 2). Figure 1B shows sessions and functional task sessions collected after the
example stimulation patterns designed so that increases second FES implant surgery.
or decreases in the percentage activation of the pattern During the virtual reality versus FES comparison
smoothly coordinated multiple electrodes to cause joint sessions, the participant cortically commanded single-joint
extension or flexion. The decoded command signal for and multi-joint movements of both a three-dimensional
controlling the mobile arm support caused no virtual arm and his FES arm to perform point-to-point
movement if it was below a certain threshold, and target acquisitions. The virtual arm (figure 2B) had no
otherwise caused abduction and adduction at a constant inertial or musculoskeletal dynamics and therefore was
rate. useful for demonstrating the quality of decoded iBCI

A
Cortical First Second B
implant FES implant FES implant

Virtual reality sessions unrelated to this study (57 h)

Virtual reality vs FES comparison sessions (18 h)

FES functional task sessions (5 h)

Study period

0 100 200 300 400

Time (days)

C D
Attempted Virtual arm Neural Attempted movement Control of virtual arm Control of FES system
Cortex Extension targets Extension targets Extension targets
movement motion features

Joint velocity
Trial

Control of
virtual arm Virtual arm
Cortex Decoder
motion
Flexion targets Flexion targets Flexion targets

Joint angle
sensor FES actuated Stimulation
Trial

Control of arm patterns

FES system
Virtual arm
Cortex Decoder 30
motion Extension targets
Flexion targets
TX rate (Hz)

20

10

1430 h 1500 h 1530 h 1600 h 0

1 0 1 2 1 0 1 2 1 0 1 2
Time Time (s) Time (s) Time (s)

Figure 2: Session overview

(A) Timeline of implants and experimental sessions. All days are referenced to the day of cortical implant (Dec 1, 2014, was day 0 after implant). (B) Example image from the virtual reality game.
The virtual arm is opaque and the target arm configuration (wrist flexion in this case) is translucent. (C) During the virtual reality versus FES comparison sessions, the participant completed
three different experimental conditions. Block diagrams of each condition and an example session timeline are shown. (D) Example raster plots showing the timing of threshold crossings (top rows)
and the average threshold crossing rates (bottom row) of a single channel tuned to wrist flexion and extension during a single-joint wrist movement task. The dotted line at t=0 indicates the
presentation of the target movement. This channel records more threshold crossings when flexion targets are presented and has similar tuning properties during all three experimental conditions.
FES=functional electrical stimulation.

4 www.thelancet.com Published online March 28, 2017 http://dx.doi.org/10.1016/S0140-6736(17)30601-3

Descargado para Anonymous User (n/a) en ClinicalKey Espanol Colombia, Ecuador & Peru Flood Relief de ClinicalKey.es por Elsevier en abril 28, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.

commands independently of any control difficulties added
by the FES-actuated arm and the stimulation patterns.
Experimental sessions consisted of three conditions
(figure 2C). During the attempted movement condition,
the participant observed the virtual arm23 make
goal-directed, point-to-point movements while he was
instructed to attempt those same movements. Neural
activity evoked during these attempted movements was
used for decoder initialisation. During the virtual reality
condition, the participant used this initial decoder to
cortically command the real-time velocities of the elbow,
wrist, hand, and shoulder joints to perform single-joint
and multi-joint movements of the opaque virtual arm to
a target configuration represented by a translucent arm
(figure 2B). Decoder parameters were updated after each
virtual reality block by recalibration of the decoder with
all available data, and then were fixed for the FES
condition. During the FES condition, the participant
performed single-joint and multi-joint movements of his
own FES-actuated arm while receiving visual feedback of
the arm movements and the target location via the virtual
reality system. The virtual arms joint angles were set
equal to the joint angles recorded by the instrumented
goniometers and the participant looked at the computer
monitor during this phase instead of his own FES-
actuated arm. This experimental set up (videos 1, 2)
enabled precise visualisation of the target arm
configurations during point-to-point FES arm
movements and facilitated comparison between virtual
reality and FES movements (since visual feedback of the
target location and the arm location was identical). To
successfully acquire a target, the participant had to
maintain the specified joint angle within a certain
tolerance around the target for 500 ms without exceeding
a maximum movement time of 812 s. The starting
position of each movement was equal to the ending
position of the previous movement. The decoder output
for the untested joints was set to zero. The appendix
(p 13) summarises all sessions and joints tested during
each day.
Functional task demonstrations
To show the potential of the system to restore meaningful
function, the participant completed both a coffee
drinking and a self-feeding task. For these sessions, we
calibrated the decoder while he observed and controlled
FES-actuated movements instead of virtual reality
movements. First, we initialised the decoder using
neural data recorded while his arm was automatically
driven by the FES system (ie, computer-controlled) to
make elbow, hand, and shoulder (mobile arm support)
movements. He was instructed to simultaneously
attempt to make the observed arm motions. We then
refined the decoder using neural data recorded while he
performed user-controlled single-joint FES-actuated
movements, cued by audio commands instead of the
motion of the virtual arm. After refinement, the neural
www.thelancet.com Published online March 28, 2017 http://dx.doi.org/10.1016/S0140-6736(17)30601-3 5
Descargado para Anonymous User (n/a) en ClinicalKey Espanol Colombia, Ecuador & Peru Flood Relief de ClinicalKey.es por Elsevier en abril 28, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
Articles
decoder was held constant for the functional tasks. We
found that this calibration scheme worked better than
calibrating with virtual reality data, potentially because
the difference in visual feedback between virtual reality
and physical reality caused the neural activity to change.
Role of funding source
The funder had no role in the experimental design,
analysis, or manuscript preparation or submission. The
funder provided funds to complete the study, including
investigator salaries, equipment costs, and research
and clinical costs. All authors had complete access to
the data. All authors authorised submission of the
manuscript, but the final submission decision was
made by the corresponding author.
Results
The study commenced on Dec 1, 2014 (day of intracortical
implant) and data are until Nov 7, 2016 (717 days after
implant). Over the course of the study, the participant
had four minor (and no serious) device-related adverse
events, all of which were treated, resolved, and reported
to the governing regulatory bodies. No additional adverse
events had occurred at the time when the last data were
collected for this publication, and the participant remains
in the BrainGate2 trial. See Online for videos
Neural activity was strongly related to the participants
intended movement commands during attempted
movement, virtual reality, and FES movement conditions.
Figure 2D shows example neural activity that was
strongly tuned to wrist flexion over wrist extension
during each condition (substantially more threshold
crossings were observed during attempted wrist flexion
over wrist extension). Similar consistency of neural
tuning between task conditions (attempted movement vs
virtual reality vs FES) on some channels was also
observed for elbow flexion or extension, humeral
abduction and adduction, and hand opening or closing
(appendix p 5). Of the 192 electrodes, we identified a
neural feature (threshold crossing or spectral high-
frequency power) that coded for hand opening or closing
on 15 electrodes (SD 2), for elbow flexion or extension
on 25 electrodes (2), for wrist flexion or extension on
25 electrodes (4), and for humeral abduction and
adduction on 27 electrodes (20; appendix p 12). The
appendix (p 11) reports the number of isolatable single
neurons recorded over time.
Across the virtual reality versus FES comparison
sessions, the participant consistently achieved 80100%
success during single-joint movements of the elbow,
wrist, hand, and mobile arm support (humeral
abduction and adduction) to specified target positions
(figure 3, video 1). For some joint movements (elbow
flexion and extension, mobile arm support abduction
and adduction), the participant acquired targets with his
FES-actuated arm as quickly and as successfully as the
virtual arm (figure 3A). For other joint movements
 Articles

A
Single joint movement Extension target Middle target Flexion target
Virtual
Elbow Extension FES

76

20
Flexion 9

Wrist Extension

39

17
Flexion
5

Hand Open

68
12
Close
7

Shoulder 12
(MAS) Up 8

40
Down

0 2 4 6 0 2 4 6 0 2 4 6
Time (s) Time (s) Time (s)

B
Single block performance Indicates shown in A Virtual reality Chance
100 50

80 45
Movement time (s)

40
Success (%)

60
35
40
30
20 25

0 20
Elbow Wrist Hand Shoulder (MAS) Elbow Wrist Hand Shoulder (MAS)

Figure 3: Single-joint FES and MAS movements under real-time brain control
(A) Restored arm and hand movements and achievable ranges of motion. Line drawings are made from actual photographs of restored movements, and show
complete range of restored motion. Overlaid time series of joint motions towards each target during an example block of each movement. Each line illustrates a single
movement from the example FES block or virtual reality block. Grey rectangles illustrate the target and tolerance allowed for target acquisition. Target distances
(from the flexion to extension target) and allowed tolerances (widths) were 434 [SD 60] for elbow, 24 [34] for wrist, 358 [58] for hand, and 413 [51] for MAS.
The participant was in full control of the joint at all times (the joint position was not reset after a target was acquired). Example blocks with high success rates are
shown. (B) Success rate and average movement time are summarised for each FES block (circles). Circles within each column are different colours if they occurred on
different days. Average virtual reality performance (blue dotted line) and chance performance (red dotted line) are shown for reference. The appendix (p 16) gives a
more detailed quantification with accompanying statistical tests. FES=functional electrical stimulation. MAS=mobile arm support.

6 www.thelancet.com Published online March 28, 2017 http://dx.doi.org/10.1016/S0140-6736(17)30601-3

Descargado para Anonymous User (n/a) en ClinicalKey Espanol Colombia, Ecuador & Peru Flood Relief de ClinicalKey.es por Elsevier en abril 28, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.

(wrist flexion or extension, hand opening or closing),
the participant achieved high success rates, but targets
were acquired more slowly and speeds varied
non-uniformly as a function of joint angle (figure 3A).
Results for two-joint and three-joint movements are
shown in the appendix (pp 6, 7) and video 2. On average,
the participant achieved fewer target acquisitions during
control of his own arm than during control of the virtual
arm (figure 3B). However, FES movements were more
successful (more target acquisitions) than chance move
ments for both single-joint and multi-joint movements,
and more successful than what he could achieve with
residual voluntary shoulder movements alone
(figure 3B, appendix pp 8, 16).
Failed reaching attempts were categorised as being due
to muscle fatigue, which made it impossible to reach the
target even at full stimulation; control interface (FES and
mobile arm support) challenges, which made it difficult
to accurately stop within the target region; or failure to
decode the correct command signal to move the joint
towards the target. The appendix (p 9) shows each failure
mode. For single-joint movements, 27 (82%) of 33 failed
trials were due to control interface challenges,
predominantly the inability to maintain a desired hand
grasp posture (appendix p 14). These trials occurred
primarily when the decoded commands for the hand
were mapped to highly non-linear portions of the
stimulation pattern. Some portions of the pattern
contained so-called dead space that did not move the
hand very much, whereas other portions caused large,
quick movements, resulting in target overshoots. We
alleviated this problem in later blocks by using an
automatic procedure that warped the stimulation pattern
so that command signals were mapped linearly to
equilibrium positions (appendix p 10). Four (12%) of
33 failed trials were due to muscle fatigue, whereas two
(6%) were due to failure to decode an appropriate
command signal (appendix p 14).
For multi-joint movements, the dominant failure mode
was also control interface challenges (contributed to 133
[63%] of 210 failed trials; appendix p 14), and was due
primarily to mobile arm support movements causing
undesired motion of other joints. Failure to decode an
appropriate command signal for at least one of the joints
was more common for multi-joint movements
(contributed to 80 [38%] of 210 failed trials; appendix
p 14) than single-joint movements. Decoding failure
might have been more common in multi-joint trials
because of the cognitive burden of controlling multiple
joints, with real dynamics and musculoskeletal
limitations of spinal cord injuries. Muscle fatigue was a
reason for failure in around 12% of multi-joint (mostly
elbow and wrist) trials.
The participant was able to successfully acquire a cup
of coffee and drink from it with a straw (figure 4A,
video 3) and feed himself using the FES+iBCI system
(video 4). The coffee drinking task required him to extend
www.thelancet.com Published online March 28, 2017 http://dx.doi.org/10.1016/S0140-6736(17)30601-3 7
Descargado para Anonymous User (n/a) en ClinicalKey Espanol Colombia, Ecuador & Peru Flood Relief de ClinicalKey.es por Elsevier en abril 28, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
Articles
his elbow, open his hand, grasp the cup securely, flex his
elbow to transport it close to his mouth, drink using the
straw, extend his elbow to return the cup, and release his
grasp. He required between 20 s and 40 s to complete the
drinking task and was successful in 11 of 12 attempts
made during the illustrated session (figure 4B). During
self-feeding, he consistently and repeatedly scooped
forkfuls of mashed potatoes, and navigated his hand to
his mouth to take several bites. When we asked the
participant to describe how he commanded the FES arm
movements, he replied, Its probably a good thing that
Im making it move without having to really concentrate
hard at it. I just think out and it just goes.
The participant was completely unable to perform
meaningful movements with the FES system turned off
(figure 4C); his minor residual shoulder girdle motion
only caused a small, uncontrolled elbow jerk and he
could not move his hand at all. This outcome indicates
that no substantial motor recovery occurred as a result of
FES or iBCI (shown more extensively in the appendix,
p 8). Video 5 shows an additional qualitative comparison
of the movements the participant could make with and
without the FES+iBCI system, and also shows that he
could move each joint individually with the FES+iBCI
system while suppressing the motion of undesired
joints. Finally, the participant required continuous visual
feedback of his arm movements, since an absence of
proprioception prevented knowledge of arm position and
therefore an inability to perform meaningful movements
without visual feedback.
Discussion
FES+iBCI restoration of arm and hand functions,
combined with a mobile arm support (standard practice
for individuals with C4 spinal cord injury implanted with
FES arm systems) under iBCI command, represents a
neurotechnology-based circumvention of spinal cord
injury. This restoration gives individuals with chronic
tetraplegia intuitive control over reaching and grasping
movements using their paralysed limbs. The study
substantially extends previous iBCI research on
individuals with paralysis controlling cursors or robotic
limbs.1013 The movements afforded to the study
participant (reaching out, grasping or scooping, reaching
back to the face) allowed him to repeatedly drink coffee
and feed himself with his own arm and hand, solely of
his own volition. These actions are representative of
movements needed to perform a wide range of reaching
tasks, which suggests that more functional activities are
achievable with the current system.
FES movements were moderately slower and less
accurate than the same movements of the virtual reality
arm under brain control. This discrepancy might have
been due to the difference in time spent practising virtual
reality versus FES (65 h vs 15 h), but might have also been
caused by the more difficult control task presented by an
FES-activated arm: dynamics due to arm mass, muscle
 Articles

I II

B
Trial phases
Reach out
Cup grasp
11
Reach in
Drink
9 Reach out
Cup release
Success
Trial number

7 Failure

1
0 10 20 30 40
Time (s)
Figure 4: Use of the FES+iBCI C
system to drink coffee
System on System off
(A) Participant reaching out to
grasp the cup of coffee (I) and Flexion
bringing it to his mouth to
drink through a straw (II).
Elbow

Photos taken on trial day 392

(Dec 28, 2015). (B) Length of
time taken to complete each
phase of the drinking task. Extension
Data are shown for 12 trials
completed within a single
experimental session.(C)
Example time series of elbow Close
and hand motion when the
FES+iBCI system was turned on
Hand

(left) and when the FES system

alone was turned off (right).
Data for (B) and (C) collected Open
on trial day 463 (March 8,
2016). FES=functional
180 190 200 15 20 25 30 35 40
electrical stimulation. Time (s) Time (s)
iBCI=intracortical Neural command Joint angle
braincomputer interface.

8 www.thelancet.com Published online March 28, 2017 http://dx.doi.org/10.1016/S0140-6736(17)30601-3

Descargado para Anonymous User (n/a) en ClinicalKey Espanol Colombia, Ecuador & Peru Flood Relief de ClinicalKey.es por Elsevier en abril 28, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.

contractile properties, interactions between joints, and
motor dynamics of the mobile arm support. In our
previous studies,24,25 we showed that these control
difficulties are addressable with a feedback control
system that converts higher level movement commands
decoded from the iBCI (eg, desired joint velocities) into
the lower level muscle stimulations needed to smoothly
achieve that movement. The feedback controller
incorporates joint angle sensors that continuously sense
any movement error or deviation from the desired
movement, and recruits the appropriate muscles to
reduce that error while taking into account the dynamics
of the musculoskeletal limb (similar to how robotic arms
are controlled). Of note, even without an implemented
feedback controller, the participant was able to modulate
his neural activity and use visual feedback alone to
perform meaningful FES arm movements, even on day 1.
The percutaneous, readily removable, FES electrodes
provide proof-of-concept for fully implanted FES
systems. Although this choice restricted the number of
joints that could be restored and their ranges of motion,
future fully implantable FES systems can take
advantage of enhanced electrode design and surgical
placement (eg, more precisely located intramuscular
electrodes implanted via open surgery, or peripheral
nerve cuff electrodes for a more distributed motor
unit recruitment)3 and associated techniques (eg,
model-based optimisation of muscle stimulation
patterns, muscle tendon transfers to replace the
functions of denervated muscles,6 and more extensive
exercise programmes) to restore motion more fully.
The use of implanted FES is crucial for clinical adoption
of this technology. Although some earlier studies have
focused on the use of surface FES to restore only hand
grasp to individuals with lower-level spinal cord injury,
either commanded by electro encephalography26,27 or
iBCIs,16 surface FES systems do not have a history of
widespread and long-term clinical adoption. By
contrast, fully implanted systems for FES grasp
restoration, specifically the Freehand,1,4 have a history of
successful clinical adoption, probably due to the
seamlessness of day-to-day setup and use, and their
durability (<1% of electrodes fail over 3 years).1
Although iBCI-commanded systems (including robotic
arms) have not yet restored movements with the same
speed and precision of able-bodied movements, the
current level of gross movement that they can restore is
still enough to achieve clinically relevant functions
(such as self-feeding). Enhanced speed, precision, and
multifunctional control might be achievable through
electrode technologies that record more neurons from
distributed cortical networks, improved decoding
algorithms, implantable FES technologies, and restored
somatosensation. Somatosensation restored through
intracortical stimulation28 might also eventually allow
users to make reach and grasp movements that are safer
in the absence of constant visual feedback, as a result of
www.thelancet.com Published online March 28, 2017 http://dx.doi.org/10.1016/S0140-6736(17)30601-3 9
Descargado para Anonymous User (n/a) en ClinicalKey Espanol Colombia, Ecuador & Peru Flood Relief de ClinicalKey.es por Elsevier en abril 28, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
Articles
sensory feedback of object properties (eg, temperature) in
the reachable workspace. Despite limitations, we believe
that iBCIs currently offer the best option for seamless
clinical integration and greater functional performance,
particularly over their non-invasive counterparts (eg,
continuous control of a high-dimensional robotic limb
has been successfully shown with iBCI systems12 but
never with electroencephalography). Research advances
in intracortical electrode biocompatibility29 and fully
implanted brain recording interfaces30 continue to
increase the clinical viability of iBCI-commanded systems.
The present FES+iBCI system offers individuals with
chronic tetraplegia from spinal cord injury the possibility
of regaining lost arm and hand function to perform
activities of daily life. Continued clinical translation
of this technology will be aided by iBCI and FES
technological advances, which will result in smoother
and more dexterous arm and hand movements. Future
systems inspired by this work might provide full-time
and more accurate control of the arm and hand, to enable
restoration of a wider range of functional activities and
resulting in increased independence and quality of life.
Contributors
ABA, PHP, JPD, LRH, and RFK had the overall conceptual design.
ABA, FRW, DRY, WDM, BAM, JDS, LRH, and RFK conceived and
designed the scientific experiments. FRW and DRY implemented the
software of the FES+iBCI system and experiments. FRW, DRY, WDM,
and BAM performed the research sessions. FRW and DRY analysed the
data. ABA, FRW, DRY, WDM, BAM, and RFK interpreted the data. JPM,
JAS, HAH, and MWK designed and implemented the surgery. BLW
provided the patient with post-surgical care. ABA, FRW, DRY, WDM,
BAM, BLW, JAS, HAH, MWK, PHP, JDS, JPD, JPM, LRH, and RFK
contributed to the writing of this Article. LRH was the investigational
device exemption sponsorinvestigator of the pilot clinical trial.
Declaration of interests
We declare no competing interests.
Acknowledgments
We thank the study participant for his pioneering efforts participating in
the present study. We thank the Louis Stokes Cleveland VA Medical
Center Cares Tower Residence Center, for space and logistical support. We
additionally thank Jaimie M Henderson and Krishna V Shenoy for their
feedback and support of the research efforts. Support for this study was
provided primarily by the Eunice Kennedy Shriver National Institute of
Child Health and Human Development (NICHD) of the National
Institutes of Health (NIH) under award number R01HD077220, and by
NICHD-NIH N01HD53403, and the National Institute on Deafness and
Other Communication Disorders of NIH under award number
R01DC009899. Support was also provided by the United States
Department of Veterans Affairs, Rehabilitation Research and
Development Service, under award numbers B4853C, B6453R, and
N9228C. ABA was partially supported by the Department of Veterans
Affairs Career Development Award B9043W. The content is solely the
responsibility of the authors and does not necessarily represent the official
views of NIH, the Department of Veterans Affairs, the United States
Government, other funding agencies, or of the parent institutions.
Caution: investigational device. Limited by federal (United States) law to
investigational use.
References
1 Peckham PH, Keith MW, Kilgore KL, et al. Efficacy of an implanted
neuroprosthesis for restoring hand grasp in tetraplegia:
a multicenter study. Arch Phys Med Rehabil 2001; 82: 138088.
2 Kilgore KL, Hoyen HA, Bryden AM, Hart RL, Keith MW,
Peckham PH. An implanted upper-extremity neuroprosthesis using
myoelectric control. J Hand Surg Am 2008; 33: 53950.
 Articles
3 Memberg WD, Polasek KH, Hart RL, et al. Implanted
neuroprosthesis for restoring arm and hand function in people with
high level tetraplegia. Arch Phys Med Rehabil 2014; 95: 120111.
4 Taylor P, Esnouf J, Hobby J. The functional impact of the Freehand
System on tetraplegic hand function. Clinical results. Spinal Cord
2002; 40: 56066.
5 Fromm B, Rupp R, Gerner HJ. The Freehand System:
an implantable neuroprosthesis for functional electrostimulation of
the upper extremity. Handchir Mikrochir Plast Chir 2001; 33: 14952
(in German).
6 Keith MW, Kilgore KL, Peckham PH, Wuolle KS, Creasey G,
Lemay M. Tendon transfers and functional electrical stimulation for
restoration of hand function in spinal cord injury. J Hand Surg Am
1996; 21: 8999.
7 Taylor DM, Tillery SIH, Schwartz AB. Direct cortical control of 3D
neuroprosthetic devices. Science 2002; 296: 182932.
8 Serruya MD, Hatsopoulos NG, Paninski L, Fellows MR,
Donoghue JP. Instant neural control of a movement signal.
Nature 2002; 416: 14142.
9 Velliste M, Perel S, Spalding MC, Whitford AS, Schwartz AB.
Cortical control of a prosthetic arm for self-feeding. Nature 2008;
453: 1098101.
10 Hochberg LR, Serruya MD, Friehs GM, et al. Neuronal ensemble
control of prosthetic devices by a human with tetraplegia.
Nature 2006; 442: 16471.
11 Hochberg LR, Bacher D, Jarosiewicz B, et al. Reach and grasp by
people with tetraplegia using a neurally controlled robotic arm.
Nature 2012; 485: 37275.
12 Collinger JL, Wodlinger B, Downey JE, et al. High-performance
neuroprosthetic control by an individual with tetraplegia.
Lancet 2013; 381: 55764.
13 Aflalo TN, Kellis S, Klaes C, et al. Decoding motor imagery from the
posterior parietal cortex of a tetraplegic human. Sci Mag 2015;
348: 90610.
14 Moritz CT, Perlmutter SI, Fetz EE. Direct control of paralysed
muscles by cortical neurons. Nature 2008; 456: 63942.
15 Ethier C, Oby ER, Bauman MJ, Miller LE. Restoration of grasp
following paralysis through brain-controlled stimulation of
muscles. Nature 2012; 485: 36871.
16 Bouton CE, Shaikhouni A, Nicholas V, et al. Restoring cortical
control of functional movement in a human with quadriplegia.
Nature 2016; 533: 24750.
17 Maynard EM, Nordhausen CT, Normann RA. The Utah intracortical
electrode array: a recording structure for potential brain-computer
interfaces. Electroencephalogr Clin Neurophysiol 1997; 102: 22839.
10 www.thelancet.com Published online March 28, 2017 http://dx.doi.org/10.1016/S0140-6736(17)30601-3
Descargado para Anonymous User (n/a) en ClinicalKey Espanol Colombia, Ecuador & Peru Flood Relief de ClinicalKey.es por Elsevier en abril 28, 2017.
Para uso personal exclusivamente. No se permiten otros usos sin autorizacin. Copyright 2017. Elsevier Inc. Todos los derechos reservados.
18 Yousry TA, Schmid UD, Alkadhi H, et al. Localization of the motor
hand area to a knob on the precentral gyrus. A new landmark. Brain
1997; 120: 14157.
19 Memberg WD, Peckham PH, Thrope GB, Keith MW, Kicher TP.
An analysis of the reliability of percutaneous intramuscular electrodes
in upper extremity FNS applications. IEEE Trans Rehabil Eng 1993;
1: 12632.
20 Kilgore KL, Peckham PH, Thrope GB, Keith MW,
Gallaher-Stone KA. Synthesis of hand grasp using functional
neuromuscular stimulation. IEEE Trans Biomed Eng 1989;
36: 76170.
21 Dickey AS, Suminski A, Amit Y, Hatsopoulos NG. Single-unit
stability using chronically implanted multielectrode arrays.
J Neurophysiol 2009; 102: 133139.
22 Koyama S, Chase SM, Whitford AS, Velliste M, Schwartz AB,
Kass RE. Comparison of brain-computer interface decoding
algorithms in open-loop and closed-loop control. J Comput Neurosci
2010; 29: 7387.
23 Ajiboye AB, Simeral JD, Donoghue JP, Hochberg LR, Kirsch RF.
Prediction of imagined single-joint movements in a person with
high-level tetraplegia. IEEE Trans Biomed Eng 2012; 59: 275565.
24 Blana D, Kirsch RF, Chadwick EK. Combined feedforward and
feedback control of a redundant, nonlinear, dynamic
musculoskeletal system. Med Biol Eng Comput 2009; 47: 53342.
25 Chadwick EK, Blana D, Simeral JD, et al. Continuous neuronal
ensemble control of simulated arm reaching by a human with
tetraplegia. J Neural Eng 2011; 8: 34003.
26 Lauer RT, Peckham PH, Kilgore KL. EEG-based control of a hand
grasp neuroprosthesis. Neuroreport 1999; 10: 176771.
27 Pfurtscheller G, Mu GR, Ju H. Thoughtcontrol of functional
electrical stimulation to restore hand grasp in a patient with
tetraplegia. Neurosci Lett 2003; 351: 3336.
28 Flesher SN, Collinger JL, Foldes ST, et al. Intracortical
microstimulation of human somatosensory cortex. Sci Transl Med
2016; 8: 361ra141.
29 Jorfi M, Skousen JL, Weder C, Capadona JR. Progress towards
biocompatible intracortical microelectrodes for neural interfacing
applications. J Neural Eng 2015; 12: 11001.
30 Capogrosso M, Milekovic T, Borton D, et al. A brain-spinal interface
alleviating gait deficits after spinal cord injury in primates.
Nature 2016; 539: 28488.