Page 378 - Handbook of Biomechatronics

372 Lilach Bareket et al.

3 CURRENT VISUAL CORTEX IMPLANT TECHNOLOGIES

Cortical visual prostheses employ electrodes that are placed in contact
with the primary visual cortex (V1 or striate cortex). The visual sensory
input that arrives from the eyes goes first through the LGN (located in
the thalamus) and then reaches V1. Each hemisphere in the visual cortex
receives information from the ipsilateral LGN. The left visual cortex receives
signals from the right visual field and vice versa. V1 is part of the occipital
lobe (in the back of the skull), that encompasses buried portions of cortex in
the calcarine sulcus and its upper and lower banks, extending posterolaterally
to the occipital pole.
A major advantage of the visual cortex as the stimulation site is the pos-
sibility to treat indications where the ganglion cells in the retina or the ON’s
are severely damaged, such as: severe retinal disease such as diabetic retinop-
athy and glaucoma, optic atrophy, optic neuritis, large pituitary or parasellar
tumors, trauma to the eyes or the ON’s, or bilateral retinoblastoma follow-
ing enucleation of both eyes. Other advantages are the relatively large area to
place electrodes compared with the retina and the LGN, which means that a
large number of electrodes can potentially be implanted. The total available
2
surface area of V1 varies between 1400 and 6300mm (depending on the
method of estimation) with approximately 67% of that area buried inside
the calcarine fissure (Andrews et al., 1997; Stensaas et al., 1974). Electrodes
can be placed on the surface of the visual cortex (subdural) or inserted into
the brain tissue (intracortical).
Two key pioneers in the development of cortical bionic vision prostheses
were Giles Brindley and William Dobelle (Brindley and Lewin, 1968;
Dobelle and Mladejovsky, 1974; Dobelle et al., 1974). By 1972, Brindley
and Lewin had implanted two devices. In the second device, enhancements
to the electronic design by Donaldson (1973) produced a “row-column”
transmission strategy which dramatically reduced the number of receivers
necessary on the subpericranial portion of the implant while maintaining
a similar quantity of electrodes (75). This reduction in receiver quantity
(and associated increase in separation between receivers) had a
corresponding reduction in “cross talk” between neighboring receivers.
This was indicated as a significant problem in the original design as transmit-
ter misalignment of as little as 5mm had caused stimulation or partial stim-
ulation of electrodes attached to neighboring receivers. The second patient
could read braille characters presented to him using the device alone at a rate