HeLPS Demo 5


Cochlear Implants

A cochlear implant is a medical device that is designed to restore a sense of hearing to otherwise completely deafened individuals. A cochlear implant stimulates the auditory nerve with a tiny electrode array, resulting in a relatively crude approximation to normal hearing. The most advanced electrode arrays today use only 22 electrodes, whereas the healthy human cochlea has thousands of hair cell receptors for transmitting information about sounds. Although surgeons attempt to implant the electrode array so that high and low-frequency sounds stimulate the parts of the auditory nerve that are normally associated with those sounds, the mapping is always very different from normal. As a result of these differences, the sense of hearing that the cochlear implant provides is fundamentally different from that of normal hearing. When the device is first turned on, the recipient is brought from a world of profound deafness to a strange world of bells, whistles, and electronic chirps. With rehabilitation and training, speech reception rapidly increases over the first two months as recipients learn to use their new auditory code.

 

Cochlear implant technology is widely regarded as a modern medical success. Over 100,000 individuals have received cochlear implants world-wide. The most successful recipients are capable of understanding speech in quiet without the use of lip-reading.

The algorithm used by HeLPS for cochlear implant simulation is shown in the upper figure.  The number of channels, N, in the simulation is an important parameter. Specifying N results in the audio band from 0.1 – 9 kHz being separated into N channels of equal width in octaves.  N can be used to adjust the quality of the signal representation and the resulting psychoacoustic performance, with a larger number of channels leading to better performance.

The processing performed within any channel of the cochlear implant simulation is shown in the inset in the figure. This simulation processing is similar to that described in the research literature (e.g., Friesen et al., 2001) in which the envelopes of bandpass filtered signals modulate noise band carriers.

There are a couple caveats about cochlear implant simulations that must be understood. First, the number of channels in the simulation is not simply related to the number of channels or electrodes in an actual implant.  The number of simulation channels only provides a means of controlling spectral resolution and, thereby, psychoacoustic performance, and should not be taken as a literal equivalent to the number of channels in an actual implant. Second, there is currently no set of parameters for characterizing an individual implant/patient combination, and customizing a simulation, in the way that audiograms and hearing aid characteristics do for acoustic hearing. All patients and implants are simulated in the same way.

There are three pages of demos, divided among speech, music, and environmental sounds. All demos present cochlear implant simulations with 1, 4, 8, 12, or 20 channels.

Friesen, L. M., Shannon, R. V., Baskent, D., and Wang, X. (2001). “Speech recognition in noise as a function of the number of spectral channels: Comparison of acoustic hearing and cochlear implants,” J. Acoust. Soc. Am. 110, 1150–1163.

Click here to open the HeLPS Cochlear Implant Demo in a new window.