Biomedical Engineering Reference
In-Depth Information
the management of a profound hearing loss (Clark, 1969). For the following 10 years he
proceeded with animal experiments. It was only in 1978 that he performed his first implant
on a human being, Rod Saunders, who had been deafened by a blow to the back of his
head in a car accident 18 months before. Saunders continued to help Clark's team for more
than 20 years, and when he died in 2007 his last wish was that his ear bones and brain be
used for further research (Carman, 2008).
Meanwhile, in 1977 the U.S. National Institutes of Health (NIH) commissioned a
study to determine whether further development of cochlear implants would be wise. One
of the conclusions of this Bilger Report (Bilger, Black et al., 1977) was that although the
subjects could not understand speech through their prostheses, they did score significantly
higher on tests of lip reading and recognition of environmental sounds. In a consensus
statement 11 years later, in 1988, the NIH suggested that multichannel implants were
more likely to be effective than single-channel implants (Wilson and Dorman, 2008). This
is because the cochlea exhibits a tonotopic organization in which the outer sections are
sensitive to low-frequency sounds with the frequency sensitivity decreasing monotonically
toward the center. At this time only about 3000 patients had received cochlear implants.
All modern cochlear implants separate the sound spectrum using band-pass filters and
use these individual outputs to stimulate different regions of the cochlea. However, new
and highly effective processing strategies were developed in the late 1980s and early 1990s
principally through the national prosthesis program in the United States. These include
continuous interleaved sampling (CIS),
m
-of-
n
, and spectral peak (SPEAK) strategies.
Large gains in speech reception performance were achieved, and CIS and
m
-of-
n
remain
in widespread use today (Wilson and Dorman, 2008).
In 1995, by which time about 10,000 patients had received implants, a second NIH
consensus development conference was held. Their primary conclusion was that a majority
of those individuals with the latest speech processors for their implants scored above 80%
correct on high-context sentences, even without visual cues.
By the middle of 1996, the cumulative number of implants exceeded 110,000, as can
be seen in Figure 6-28. This is orders of magnitude higher than the numbers for all other
neural prostheses, including those for restoration of motor and other sensory functions.
Contemporary manufacturers include AllHear Inc. (Aurora, Oregon), Clarion
®
(Advanced Bionics, Inc., Sylmar, California), Nucleus
®
(Cochlear Corporation, Sydney,
Australia), Digisonic
®
(MXM Co., Vallauris, France), Interaid (Symbion, Inc., Provo,
Utah), Laura Flex (Antwerp Bionic Systems, Belgium), and COMBI
TM
-40
(MED-EL
Corp., Insbruck, Austria). In addition, a number of research institutes and universities are
also pursuing improved electrode arrays and better signal-processing algorithms.
+
6.9.2 How Cochlear Implants Work
As shown in Figure 6-29, a modern cochlear implant consists of a microphone, external
sound processor and power supply, transmitting circuitry, the receiver-stimulator package,
and an electrode array.
The microphone picks up sounds, and the signal processor filters and selects the
information and converts it into electrical signals that are transmitted to the intracochlear
electrode array. Both the encoded signal and the power are transmitted transcutaneously,
using radio frequency antennas, to a demodulator that assigns speech information to the
electrode array. Single channel systems use only one electrode, while multichannel systems
employ up to 31 channels.
