Brain-computer interfaces (BCIs) are my research area, so I'm used to the
terminology.
P300 is the type of brain activity and could therefore theoretically be
picked up using ECoG (implanted electrodes) or EEG (scalp electrodes),
though I can't think of an example where ECoG has been used to record
P300.
ECoG refers to the type of electrodes used for the recording. There are
generally three kinds: electroencephalograph (EEG) from the scalp,
entirely non-invasive; electrocorticogram (ECoG) from electrodes implanted
inside the skull, but not penetrating the brain, which is sometimes
referred to as minimally invasive; and spikes trains or single cell
recordings from microelectrodes which are implanted into the brain and
penetrate brain tissue. Currently, EEG is widely used by neurologists,
ECoG is also widely used, including for implants of some length and
microelectrodes are being researched, though they are increasingly in
trials with human subjects.
The sentence "The device under development also carries less risk than
BCIs that surgically implant electrodes in the brain" is (I think)
referring to the common thought among ECoG BCI researchers that implanted
electrodes inside the skull, but not into the brain itself is less risky
than sticking them into the brain tissue. While I can't argue with that
point, I can argue that they aren't very clear about what they are
actually proposing. I find that the result of statements like that is
often to mis-lead the reader.
There are two companies that are commercially developing microelectrodes
and testing them in human subjects. Neurosignals, Inc and
Cyberkinetics. They've each done a half dozen or so implants in humans.
Neurolutions is the first ECoG company I've heard of.
I don't know of a EEG BCI company (which is strange, since it is the
farthest along in my opinion). I also haven't heard of a company working
on a commercial P300 interface. You can buy the hardware for a P300 BCI,
though the pricetag will likely be $15-20K. The BCI2000 software you can
likely get for free (if you do enough paperwork), but there isn't much
technical support available. Still, we were able to get it working in our
lab after about three weeks of fiddling with it. However, it isn't a
commercial product yet and there are lots of usability and support issues
that remain.
My current research is focused on removing the barriers to commercial
availability of an EEG BCI. However, I don't have the energy to start a
company and I don't know of anyone who is currently focused on
commercialization of a P300 interface. There are practical issues that
need to be solved before a company could be a commercial success. I'm
working on identifying them and finding ways around them so that
commercialization can proceed. One could need to evaluate the amount of
training and technical support needed as well as other functional issues
regarding what the BCI does when you don't feel like typing. Interfacing
BCIs with commercial assistive technology is one of my projects and we now
have a prototype connector for the BCI2000 that will allow it to be
recognized as a USB keyboard by just about anything. One part of the
experiment we're working on at the moment is using it to type into a
DynaWrite. Anyway, I can write about this for pages and pages, so I'd
better stop now. You can check out my project's webpage or contact me
off-list if you're interested in more details.
http://www.engin.umich.edu/dbi/
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Jane Huggins, Ph.D. "If, through some inscrutable act of providence, I
were to lose all my faculties save one, I would
choose to retain the gift of speech, for through it
xxxxxx@xxxxxxxxx I would soon regain all the rest."-- Daniel Webster
%%%%%%%%%%%%%%%%% http://www-personal.umich.edu/~janeh %%%%%%%%%%%%%%%%%
On Thu, 26 Mar 2009, Edward Hitchcock wrote:
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">Thanks for reading more closely than I initially did!
Am I correct in thinking that P300 signal is distinct from ECoG that the
article is referencing? Then we really are talking about 2 different
systems in any case.
But the last sentence seems to reference that this system is not
implanted on the cortex. "The device under development also carries
less risk than BCIs that surgically implant electrodes in the brain".
Do you know if the P300 is any closer to being commercially available?
My understanding was that one could purchase a system, but overall, it
was still pretty raw. I talked with them a little bit about doing some
work on it to interface it with better AAC strategies than one character
at a time.
Thanks Jane!
Ed Hitchcock OTR/L
Technology Center
Rehabilitation Institute of Chicago
-----Original Message-----
From: xxxxxx@xxxxxxxxxxxxxxxxxxxx
[mailto:xxxxxx@xxxxxxxxxxxxxxxxxxxx] On Behalf Of Jane Huggins
Sent: Thursday, March 26, 2009 1:30 PM
To: xxxxxx@xxxxxxxxxxxxxxxxxxxx
Subject: RE: [Assistive Technology] BCI Licensed to Private Firm --
Commercial Product on the Horizon?
I'm aware of the P300 interface for typing using BCI2000 that was
demonstrated at RESNA last year. I'm actually using it in my lab right
now. It is very impressive, though its top typing speed is still
measured
in characters per minute and usually single digits at that. But, it
works
and it is high time someone commercialized it.
However, I don't think from the article that this is what Neurolutions
is
working on. It specifically states that their device uses ECoG and all
other references to Neurolutions that I can find on the web talk about
an
implanted system. While the last sentence of paragraph 4 does reference
scalp electrodes, it does so as a comparison with the claim that the
device they're developing requires less training than scalp based
systems.
So, I don't think Neurolutions is developing a P300 interface as a
commercial product. Also, all previous BCI work from Washington
University in St. Louis has been ECoG work (and good stuff at that).
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%
Jane Huggins, Ph.D. "If, through some inscrutable act of providence, I
were to lose all my faculties save one, I would
choose to retain the gift of speech, for through
it
xxxxxx@xxxxxxxxx I would soon regain all the rest."-- Daniel
Webster
%%%%%%%%%%%%%%%%% http://www-personal.umich.edu/~janeh
%%%%%%%%%%%%%%%%%
On Thu, 26 Mar 2009, Edward Hitchcock wrote:
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">Actually, I saw these guys at RESNA last year and was very impressed.
This particular system does NOT implant directly on the cortex, it is
all external scalp stuff. (The article cites that as well, last
sentence of 4th paragraph).
The system I saw was not being used to control cursor or isolated
movement. I hesitate to try to describe via email, as it is not
</blockquote>really
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">any kind of scanning either. But it was reliable for slow production
</blockquote>of
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">text, and they had it connected with WordQ to speed it up a little.
Basically it works that when the desired letter is highlighted, (and
they are highlighted in an apparently random pattern; it is not
</blockquote>scanning
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">as we think of it at all) the client responds with an "Ah hah"
</blockquote>thought.
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">Since the Ah hah thought brainwave always occurs at a reliable time
following the presentation of the highlighted letter, the computer
</blockquote>would
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">know that was the intended letter.
The one I saw was not using motor control of any kind (meaning no
</blockquote>motor
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">cortex) and I am probably doing a very poor job of describing this.
</blockquote>(I
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">want to say the signal was called P300).
Probably only beneficial for very severely locked in folks, past
</blockquote>eyegaze
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">and any form of switch. I think a reasonable fast switch scanner
</blockquote>would
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">still be faster than this system as well.
But I was positively impressed with it for our folks with no switch
access.
Ed Hitchcock OTR/L
Technology Center
Rehabilitation Institute of Chicago
-----Original Message-----
From: xxxxxx@xxxxxxxxxxxxxxxxxxxx
[mailto:xxxxxx@xxxxxxxxxxxxxxxxxxxx] On Behalf Of Jane Huggins
Sent: Thursday, March 26, 2009 12:41 PM
To: xxxxxx@xxxxxxxxxxxxxxxxxxxx
Subject: Re: [Assistive Technology] BCI Liscenced to Private Firm --
Commerical Product on the Horizon?
Interesting. This is the first I've heard of this. But, everyone
should
note that the proposed BCI is using ECoG as it's input. The article
doesn't do a very good job of explaining that ECoG comes from
</blockquote>electrodes
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">
implanted inside the skull but not penetrating the brain (usually the
electrodes are placed subdurally, ie under the membranes covering the
brain). I worked with ECoG for many years and think that ECoG will
provide the best signals for BCIs. However, they require a fairly
</blockquote>major
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">
surgery to implant. So, while this is an exciting step, I wanted to
make
sure you realize the model of the system they're proposing.
</blockquote>%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">%%
Jane Huggins, Ph.D. "If, through some inscrutable act of providence,
</blockquote>I
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em"> were to lose all my faculties save one, I would
choose to retain the gift of speech, for
</blockquote>through
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">it
xxxxxx@xxxxxxxxx I would soon regain all the rest."-- Daniel
Webster
%%%%%%%%%%%%%%%%% http://www-personal.umich.edu/~janeh
%%%%%%%%%%%%%%%%%
On Thu, 26 Mar 2009, Antoinette Verdone wrote:
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">Anyone have any further info on this?
</blockquote>
</blockquote>http://www.health.state.ny.us/press/releases/2009/2009-03-25_brain_comp_
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em"><blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">interface_tech_neurolutions.htm
Antoinette Verdone, MSBME, ATP
Assistive Technology Specialist
The ALS Association, Greater New York Chapter
NEW ADDRESS:
42 Broadway, Suite 1724
New York, NY 10004
Phone: 212-720-3054
Fax: 212-619-7409
Email: xxxxxx@xxxxxxxxxx
www.als-ny.org
"One cannot consent to creep when one has the impulse to soar" --
</blockquote>Helen
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">Keller
</blockquote>This e-mail, and any attachments, is intended exclusively for the
</blockquote>individual or entity to which it is addressed and may contain legally
privileged and/or confidential information, including but not limited to
protected health information.
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">If the information contains legally privileged or confidential
</blockquote>information, you have an obligation to comply with all laws and
regulations regarding its disclosure and safe keeping.
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">If you are not the intended recipient of this e-mail, you are hereby
</blockquote>notified that any dissemination, distribution, printing or copying of
this e-mail, and any attachments, is strictly prohibited.
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">If you have received this e-mail in error, please notify the sender
</blockquote>immediately and destroy the material in its entirety, whether in
electronic or hard copy format.
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">If you have a concern related to the receipt or disclosure of such
</blockquote>information, please do not hesitate to contact RIC's Privacy Officer at
312.238.7066 or xxxxxx@xxxxxxxx
<blockquote style="border-left: #5555EE solid 0.2em; margin: 0em; padding-left: 0.85em">Thank you.
</blockquote>This e-mail, and any attachments, is intended exclusively for the individual or entity to which it is addressed and may contain legally privileged and/or confidential information, including but not limited to protected health information.
If the information contains legally privileged or confidential information, you have an obligation to comply with all laws and regulations regarding its disclosure and safe keeping.
If you are not the intended recipient of this e-mail, you are hereby notified that any dissemination, distribution, printing or copying of this e-mail, and any attachments, is strictly prohibited.
If you have received this e-mail in error, please notify the sender immediately and destroy the material in its entirety, whether in electronic or hard copy format.
If you have a concern related to the receipt or disclosure of such information, please do not hesitate to contact RIC's Privacy Officer at 312.238.7066 or xxxxxx@xxxxxxxx
Thank you.
</blockquote>
