Interfacing brains with machines: public engagement and potential to benefit human health and quality of life
Submitting Institution
University of ReadingUnit of Assessment
Electrical and Electronic Engineering, Metallurgy and MaterialsSummary Impact Type
TechnologicalResearch Subject Area(s)
Information and Computing Sciences: Artificial Intelligence and Image Processing
Medical and Health Sciences: Neurosciences
Summary of the impact
The Cybernetics team at the University of Reading works at the frontier
of human-machine interaction. The group carries out research on therapy
and human enhancement in collaboration with medical professionals, to
research new therapeutic treatments for patients with paralysis. Our work
has led to the first human implantation of BrainGate, an intelligent deep
brain stimulator, and the culturing of neurons within a robot body. Our
work has been used by neurosurgeons in experimental human trials with the
aim to enhance the standard of living of paralysed individuals. This
ground breaking, and sometimes controversial work, has sparked widespread
discussion and debate in the public sphere, within the media and at the
government level, on the use of machines to enhance humans and vice versa.
Underpinning research
The BrainGate is a 100-electrode implant developed by Cyberkinetics, Inc.
(originally a Brown University spin-off) that links the nervous system to
a computer network, enabling the nervous system to be stimulated
externally. The Reading team used this implant, together with technology
developed at the Unit [1] (operational system, signal processing,
communications), to achieve the first human implant of BrainGate. The
implant was fired into the median nerves of Professor Warwick's left arm
on March 14th 2002 in a 2 hour operation. The implant was
connected to a processing unit to allow stimulation of nerve fibers to
artificially generate sensation perceivable by the subject and recording
of local nerve activity to form control commands for wirelessly connected
devices.
During evaluation of the implant system the nervous system of Professor
Warwick was connected onto the internet in Columbia
University, New York, enabling a robot arm in the University of
Reading to use Professor Warwick' neural signals to mimic his hand
movements, while allowing him to perceive what the robot touched from
sensors in the robot's finger tips [2]. Further studies also demonstrated
that it was possible to communicate directly between the nervous systems
of two individuals, the first direct and purely electronic communication
between the nervous systems of two humans
Further research, from 2004 to present, with MRC support, used an
artificial intelligence system connected to the brain via deep embedded
electrodes to accurately predict the onset of Parkinsonian tremors up to
20 seconds in advance [3,4]. The electrodes were then used to provide deep
brain stimulation to counteract these tremors before they took hold. This
is in direct contrast to traditional deep brain stimulation techniques
where a continuous stimulation signal is used to control tremors and
therefore this research is opening up other possible routes to treatment.
Finally, from 2007 to present, with various EPSRC listed support, we
cultured a population of approximately 100,000 to 150,000 living neuronal
cells (rat and human) on a multi-electrode array within a closed loop
robot body. In this novel set-up the cultured brain cells were connected
within a robot body [5]. The robot's sensory readings were used to
stimulate the cell culture and the cell activity was then in turn used to
change the robot's behaviour — for example, to control its movement inside
a small corral. The work demonstrated that cultures of neurons can be used
to control a robot to avoid obstacles, thereby creating a platform for
investigating culture plasticity (the capacity for neural pathways to
change due to feedback) [6].
These experiments have led to a basic reassessment of how the brain and
nervous system operate. The work could lead to new therapies to directly
help those with a disability or neurodegenerative disease such as
Parkinson's. The work reveals the tantalising, but for some, provocative,
possibility of being able to change and possibly enhance the basic
functioning of a brain by linking it directly with technology.
The research was led by Professor Kevin Warwick who joined the University
as Professor of Cybernetics in 1988. Professor Nasuto (Reader in
Cybernetics since 2000, becoming Professor in 2012) and Dr Mark Gasson
(research staff since 2004) formed part of the team developing this
technology. Dr Ben Whalley joined the department of Pharmacy in 2005 and
has collaborated with Professor Warrick's team on the biomedical aspect of
this research.
Other external biomedical collaborators include consultant neurosurgeons
Tipu Aziz, Peter Teddy and Amjad Shad at the John Radcliffe Hospital,
Oxford, UK.
References to the research
Key publications:
1. *M. Gasson et al., "Invasive Neural Prosthesis for Neural
Signal Detection and Nerve Stimulation", Int. J. Adaptive Control
& Signal Processing, 19, 365-375, (2005).
2. *K. Warwick et al., "The Application of Implant Technology for
Cybernetic Systems", Archives of Neurology 60, 1369-1373
(2003).
3. *D. Wu et al., "Prediction of Parkinson's Disease Tremor Onset
using Radial Basis Function Neural Network Based on Particle Swarm
Optimization", International Journal of Neural Systems 20,
109-118 (2010).
4. J. Burgess et al., "Identifying Tremor-Related Characteristics
of Basal Ganglia Nuclei During Movement in the Parkinsonian Patient", Parkinsonism
& Related Disorders, 16, 671-675, (2010).
5. K. Warwick et al., "Controlling a Mobile Robot with a
Biological Brain", Defence Science Journal, 60, 5-14,
(2010).
6. D. Xydas et al., "Revealing ensemble state transition patterns
in multi-electrode neuronal recordings using hidden Markov models", IEEE
Transactions on Neural Systems & Rehabilitation Engineering 19,
345-355 (2011).
The research has been published in high quality peer-reviewed journals
and has been internally assessed as of at least 2* quality. Those
suggested for quality assessment as indicated with *.
Key grants:
K. Warwick, S. Nasuto and V. Becerra, "Investigating the computational
capacity of cultured neuronal networks", EPSRC, 2007-2010,
£491,000.
K. Warwick, "Toward the prediction of Parkinson Disease tremor for a
demand-driven deep brain stimulator", MRC, 2007, £67,409.
S. J. Nasuto (jointly with University of Plymouth), "Brain-computer
interface for monitoring and inducing affective states", EPSRC,
2011-2016, £1,150M (University of Reading share £614,000).
S. J. Nasuto (jointly with Goldsmiths College), "Cognition as
communication and interaction", Sir John Templeton Foundation,
2011-2013, £120,000 (University of Reading share £60,000).
S. J. Nasuto, "NeuroCloud: developing a hybrid cloud architecture for
neuroscience research", EPSRC, 2010-2012, £250,000
(co-investigator)
S. J Nasuto, "Towards an integrated neural field computational model of
the brain", EPSRC, 2010, £248,000 (co-investigator grant).
S. J. Nasuto, "Bridging the gaps initiative, complex cognitive science",
EPSRC, 2007-2010, £556,000 (co-investigator grant).
Details of the impact
Impact on ethical discussions and public engagement
This research has garnered considerable public interest through magazine
articles, radio and television pieces (international news, discussion and
documentaries), e.g. cover story on Wired Magazine http://www.wired.com/wired/archive/8.02/warwick.html,
Late Night with Conan O'Brien (NBC), Ideas that Changed the World (BBC)6
in 2010, YouTube clips, blogs and Wikipedia discussions. Warwick has been
featured in multiple radio and news outlets, and popular science
television programmes discussing this research, e.g. Museum of Curiosity
on Radio 4 in 2013. The work has been presented at various prestigious
lectures, e,g. IET
Pinkerton Lecture in 2012 World
Science Festival in New York City 2009, schools, TEDx talks, e.g.
University of Chicago
in 2011, University
of Warwick in 2012, and through invited talks at medical,
philosophical and technical conferences. An exhibition in London's Science
Museum was devoted to the work using cultured neurons in a robotic
machine, and was in place for 18 months from 2008. The Science Museum
maintains a website specifically dedicated to our research7, a
resource used by schools and colleges as an integral part of their course
structure. The research also featured in the 2012 Wellcome Trust (London)
"Superhuman" exhibition, http://www.wellcomecollection.org/press/press-releases/superhuman.aspx,
which was visited by nearly 80,000 people1. The work has been
subject to many schools projects and presentations globally, some under
the auspices of the Royal Society via its Acclaim project, others due to
local organisation. In 2011, Google selected the implant research as a
major source of inspiration for young scientists. The work was used as one
of only five examples in the world for the Google Science Fair8.
The collaboration with Dr Ben Whalley integrating cultured neurons into a
robot body appeared as a story and video in New Scientist2. The
two versions of the video have collectively been viewed over 2.5 million
times9. The work was also included in the Faraday Schools
programme13in 2013.
Impact on human health and quality of life
As a direct consequence of the first successful BrainGate implant in a
human, studies have subsequently been performed to assist individuals who
are paralysed to considerably enhance their quality of life4,12.
The research carried out at Reading showed that the concept of
technological integration with the human nervous system was possible, the
Unit pioneered the first direct human neural control of a robotic arm, its
success and lack of adverse effects contributed to the obtaining of
ethical approval for further implantations of the BrainGate system4.
The subsequent use of BrainGate has improved patient outcomes, as
demonstrated in human trials reported in Nature in 201212.
BrainGate has enabled paralysed individuals to control technology around
them directly from their neural signals. Research on the use of BrainGate
has now moved on to a formal human clinical trial (trial identifier:
www.clinicaltrials.gov/ct2/show/NCT00912041). In 2013 the
Director of the Neurorehabilitation Engineering Center at the University
Medical Center Gottingen, NL used a similar approach for prosthesis
control by neural signals and specifically referred to Reading's work in
his design11.
In 2009 the Director of the Alfred Mann Institute for Biomedical Eng.,
USA, integrated Reading's research with biocompatible tech and
specifically referred back to the Unit's work10
Research into the intelligent deep brain stimulator is on going2
although in a Mail on Sunday article of 2008 the design of an intelligent
on demand stimulator by the team was hailed as "the most significant
recent advance in biomedical engineering"14.
Impact on government awareness of technology developments and their
implications
In (2010)The
Danish Council of Ethics used the Unit's research to question the
ethical aspects of cyborg technology, leading to the release of a suite of
recommendations http://bit.ly/1e7tO4d
At the UK government's Future Horizon Scanning Meeting in 2009, the
University of Reading team was specifically involved5 in
developing the government's stance on national security. This included
issues such as identity, neural control and communication as considered in
our research. As a consequence the Reading team is also involved in the
planning and presentation of the forward looking "Cross-Government Futures
Symposium" to be held in Feb 2014. The Symposium is designed to consider
differing views and perspectives about the future and what they will mean
for UK Government strategy and policy beyond the next Parliamentary term5.
In terms of these issues the group's direct, practical work across the
spectrum of neural implants and growing brains, as described here, is of
direct relevance in shaping future Government policy.
Sources to corroborate the impact
- Curator of Temporary Exhibitions at the Wellcome Trust, London (*)
- Senior Technology Correspondent, New Scientist (*)
- Consultant Neurosurgeon at the John Radcliffe Hospital, Oxford and
Professor of Neurosurgery at Oxford University - Parkinson's Disease
applications (*)
- Vascular and Critical Care Neurologist at Massachusetts General
Hospital, Boston and Visiting Associate Professor of Neurology at
Harvard University - BrainGate (*)
- Defence Science and Technology Laboratory, Horizon Scanning Team
Leader (*)
- http://www.imdb.com/title/tt1791025/
- http://www.sciencemuseum.org.uk/antenna/ratbrains/
- http://www.youtube.com/watch?v=-lqNO3hij_E
- http://www.youtube.com/watch?v=1-0eZytv6Qk
- J. Schulman, "Brain Control and Sensing of Artificial Limbs", in
Biological & Medical Physics, pp. 275-291, Springer, 2009.
- D. Farina et. al., "Controlling Prostheses using PNS Invasive
Interfaces for Amputees", Chapter in Introduction to Neural Engineering
for Motor Rehabilitation, Wiley, 2013.
- L. Hochberg et al., "Reach and Grasp by people with Tetraplegia using
a Neurally Controlled Robotic Arm", Nature, 485, pp.372-375, 2012.
- http://www.faradayschools.com/site-map/events/
- https://sites.google.com/site/lmssonline/lmss-blog/professorkevinwarrick13thjanuary2011
(*) Contact details provided separately as per guidance