Page 56 - Spring2019

The Remarkable Cochlear Implant
A Snapshot of the History
The courage of the pioneers made the
modern CI possible. They persevered
in the face of vociferous criticism, and
foremost among them was William F.
House, MD, DDS, who with engineer
Jack Urban and others developed devic-
es in the late 1960s and early 1970s that
could be used by patients in their daily
lives outside the laboratory. Addition-
ally, the devices provided an awareness
of environmental sounds, were a help-
ful adjunct to lipreading, and provided
limited recognition of speech with the
restored hearing alone in rare cases. “Dr.
Bill” also developed surgical approaches
for placing the CI safely in the cochlea
and multiple other surgical innovations,
described in his inspiring book (House,
2011). House took most of the arrows
from the critics and without his perse-
verance, the development of the mod-
ern CI would have been greatly delayed
if not abandoned. He is universally ac-
knowledged as the “Father of Neurotol-
ogy,” and his towering contributions are
lovingly recalled by Laurie S. Eisenberg
(2015), who worked closely with him beginning in 1976 and for well over a decade thereafter and stayed in touch with him until his death in 2012.
In my view, five large steps forward led to the devices and treatment modalities we have today. Those steps are
(1) proof-of-concept demonstrations that a variety of audi- tory sensations could be elicited with electrical stimulation of the auditory nerve in deaf persons;
(2) the development of devices that were safe and could function reliably for many years;
(3) the development of devices that could provide multiple sites of stimulation in the cochlea to take advantage of the tonotopic (frequency) organization of the cochlea and as- cending auditory pathways in the brain;
(4) the discovery and development of processing strategies that utilized the multiple sites far better than before; and (5) stimulation in addition to that provided by a unilat- eral CI, with an additional CI on the opposite side or with acoustic stimulation in conjunction with the unilateral CI.
This list is adapted from a list presented by Wilson (2015).
54 | Acoustics Today | Spring 2019
Figure 1. Block diagram of the continuous interleaved sampling (CIS) processing strategy for cochlear implants. Circles with “x,” multiplier blocks; green lines, car- rier waveforms. Band envelopes can be derived in multiple ways and only one way is shown. Inset: X-ray image of the implanted cochlea showing the electrode array in the scala tympani. Each channel of processing includes a band-pass filter (BPF); an envelope detector, implemented here with a rectifier (Rect.) followed by a low-pass filter (LPF); a nonlinear mapping function, and the multiplier. The output of each channel is directed to intracochlear electrodes, EL-1 through EL-n, where n is the number of channels. The channel inputs are preceded by a high-pass preemphasis filter (Pre-emp.) to attenuate the strong components at low frequencies in speech, music, and other sounds. Block diagram modified from Wilson et al. (1991), with permission; inset from Hüttenbrink et al. (2002), with permission.
Step 1 was taken by scientist André Djourno and physician Charle Eyriès working together in Paris in 1957 (Seitz, 2002) and step 5 was taken by Christoph von Ilberg in Frankfurt, Joachim Müller in Würzburg, and others in the late 1990s and early 2000s (von Ilberg et al., 1999; Müller et al., 2002; Wilson and Dorman, 2008). Bill House was primarily re- sponsible for step 2, and the first implant operation per- formed by him was in 1961. Much more information about the history is given by Wilson and Dorman (2008, 2018a), Zeng et al. (2008), and Zeng and Canlon (2015).
A Breakthrough Processing Strategy
Among the five steps, members of the Acoustical Society of America (ASA) may be most interested in step 4, the discov- ery and development of highly effective processing strategies. A block diagram of the first of those strategies and the pro- genitor of many of the strategies that followed, is presented in Figure 1. The strategy is disarmingly simple and is much simpler than most of its predecessors that included complex analyses of the input sounds to extract and then represent se- lected features of speech sounds that were judged to be most