Hi Listers,

I received a few private responses to the article on "Artificial
Vision."   This caused me to think that some of you might be interested
in this follow-up posting from another list.

We're fortunate to have Dr. Dagnelie, from Johns Hopkins, as a regular
contributor to the RPlist.  The following post was in response to the
subject of the article posted earlier on the VICUG-L list, called
"Artificial Vision.":
   ~~~~~~~~~~~~~~~~~~~~~~~
Subject: Artificial Vision Developments Today
----- Original Message -----
From: Gislin Dagnelie <[log in to unmask]>
To: <[log in to unmask]>
Sent: Monday, January 17, 2000 4:39 PM
Subject: Re: [RPLIST] Artificial Vision Developments Today


Hello Will and list:

 You may remember, from
some of my notes about the intraocular prosthesis, that this cortical
approach has always been considered the only one for those who go
blind from a degeneration or trauma that involves the inner retinal cell
layer (the ganglion cells) and/or the optic nerve.  For these people,
this is the first sign of real progress in 20 years.

Back in 1979, I visited Dr Dobelle in New York.  I was a graduate
student in Amsterdam at the time, and knew of the experiments by
Brindley in England in the late 1960s, and later by Dobelle in the
early 70s.  They had demonstrated, independently of eachother, that
implanted electrodes over the visual cortex can be used to elicit
small phosphenes in people who have become blind later in life.
   When I visited Dr Dobelle, he was rather reticent about his plans,
although he did tell me about a recent implantation with 68 platinum
electrodes: the same man who now, more than 20 years later, and
with the 4th or 5th generation of computer providing the "images," is
still "seeing" with the same electrodes.  I'm quite impressed that
these electrodes still work, after all this time.

At the same time, though, I have to add that a combination of 20/400
with a very small island of vision is not great vision:  "Jerry" cannot
fit
a whole letter in his small field and at the same time tell what letter
it
is.  You can see this for yourself, if you go to Dr Dobelle's website:

http://www.artificialvision.com

and then link to the bullet "Video clips" on the list under the
schematic drawing.  Choose the video clip called "computer.avi," and
you'll see how the patient, with his small visual field, scans the
letter
"E" little by little, to determine its orientation.

One of the expectations for cortical prostheses is that they will be
able to provide quite sharp central vision; close to 20/20 in fact,
because the projection of the visual field center onto the cortex is
very large compared to that of the periphery.  Conversely, because
the projection of the peripheral field is buried in the folds of the
brain,
off the medial wall (i.e., the part of the cortex where the two halves
of
the brain face eachother), it is almost impossible, with current
means, to convey any phosphenes away from the center of fixation,
which explains the tunnel vision.

The reason for "Jerry's" poor resolution is, probably, the type of
electrodes used.  These platinum wires were placed at the surface of
the cortex.  This was good froom the point of view of durability: They
did not damage cells in the underlying brain tissue.  But what almost
certainly happened is something I'm all too familiar with from my own
Ph.D. work with implanted electrodes over the visual cortex of trained
rhesus monkeys, and which you'll hear from anyone who has ever
worked with such electrodes (in humans just as in animals):  The
electrodes get encapsulated by connective tissue, and this creates a
semi-insulating layer as well as distance between the electrodes and
the cells they are stimulating.  Because of this, the electrodes have
to be stimulated more forcefully, and in turn this activates many cells
over a larger area.  Rather than a small point, each phosphene
becomes a larger blob, making it harder to distinguish them, and
making the composite picture more fuzzy.  This explains the poor
resolution.

By the way, "Jerry" has both a video camera (over his right spectacle
glass) and a laser range finder (over his left spectacle glass).. The
video camera provides the images that go to the electrodes, but my
understanding is that the range finder is used to eliminate from the
image any information that is either too close or too far to be of
interest.

There is one point where the USA Today article, or rather Dr Dobelle,
is not correct:  The distribution of the phosphenes is probably very
irregular, so the comparison with a stadium scoreboard (which is a
good one for the regular phosphene array one can obtain with a
retinal prosthesis) is overly optimistic.

All in all, it appears to me that Dr Dobelle has made good progress in
the image processing and stimulation hardware, but has not
demonstrated much progress in the electrodes he is using.  This is
the strongest side of Dr Normann's work in Utah, and I hope that Dr
Dobelle will be wise enough to make use of that knowledge.

What is up next, in my estimate, is a new series of implantations in
Switzerland, where Dr Dobelle runs a parallel institute, which as far
as I can tell is a commercial enterprise.  FDA approval, as stated in
the USA Today article, will probably take quite a while, although
again Dr Dobelle could make some headway by joining forces with
the electrode developments at either NINDS or the Univ of Utah.

Hope this helps,

Gislin

Gislin Dagnelie, Ph.D.
Assistant Professor of Ophthalmology
Lions Vision Research & Rehab Center
Johns Hopkins Univ. Sch. of Medicine
550 N. Broadway, 6th floor
Baltimore, MD 21205-2020      USA


VICUG-L is the Visually Impaired Computer User Group List.
To join or leave the list, send a message to
[log in to unmask]  In the body of the message, simply type
"subscribe vicug-l" or "unsubscribe vicug-l" without the quotations.
 VICUG-L is archived on the World Wide Web at
http://maelstrom.stjohns.edu/archives/vicug-l.html