Intensity Coding in Cochlear Implants

-
+
-
+
-
+
-
+
-
+
v
-
+
-
+
nanoC
Pure Tone
Speech
White Noise
 
Hide Active
Remember
Clear All
 
Peaks (+/-)
-
+
-
+
-
+
-
+
 

Introduction

The Otoconsult ICCI (Intensity Coding in Cochlear Implanets) application is an interactive graphical representation of the intensity coding-function (ICF) in cochlear implant systems. The bottom graph depicts the behavior of the CI system as a whole. It plots the system's output level in function of an input signal's intensity. The top graphs are 2 decomposed steps:

Input Signals

The ICF is plotted in response to any of 3 types of signals: a pure tone (adjustable in frequency between 100 and 8000 Hz), a speech signal or a white noise signal. It is thereby assumed that: (i) the frequency of the pure tone signal equals the center frequency of the observed channel's filter band, and as such completely falls into that channel (there is 0 dB attenuation with regard to the broadband energy); (ii) the speech signal does not fall into a single channel, it is distributed over multiple channels in such a way that the observed channel receives energy from the speech signal that is equal to the broadband energy attenuated by 12 dB; (iii) the white noise signal is distributed over all channels such that the observed channel receives energy from the white noise signal that is equal to the broadband energy attenuated by 24 dB. The attenuation of 24 dB relates to the fact that the average CI channel has a bandwidth of 493 Hz which is 1/16 of the total bandwidth (8 kHz) of current generation systems. For the speech signal, having at any given moment a band width between those of a pure tone and white noise, the attenuation inflicted by the filter bank was chosen to be the mean of the attenuations of the pure tone (0 dB) and the white noise (24 dB) signals, hence 12dB. It must be noted that the choice of these attenuations is a simplification and that in reality the attenuation is highly dependent on how the channel band pass filter is organized in relation to the input signal. It is also assumed that all input signals feature a stable long term intensity by which AGC systems reach convergence (i.e. broadband intensity is maintained stable for a time longer than the system's attack time). This response to long term intensities incorporates, amongst others, the static gain function of AGC systems (a new gain is determined from this function when the long term intensity of the input signal changes). Nonetheless the ICF plots in ICCI allow showing the response to rapid fluctuations (open symbols) around this long term intensity (filled symbols), of which it is assumed that they do not trigger the slow detectors of AGC systems.

Microphone

In the ICF plots the front end preprocessing and filter bank steps are combined and depicted on the right chart. The microphone's frequency response is not considered (i.e. it is assumed to be flat). The system noise (primarily determined by the microphone noise floor) is assumed to be white within the range being processed (0.1 to 8 kHz) and equivalent to 35 dB SPL within an acoustical band of that same range. As with the white noise input signal, the energy per channel is assumed to be equal to the broadband energy attenuated by 24 dB, and would therefore be approximately 11 dB SPL per channel on average. For all devices, the pre-emphasis filter is assumed to be an A-weighting filter. The ICCI application allows adjusting the observed channel's centre frequency, such that the effect of pre-emphasis becomes apparent.

Mapping

The mapping step is depicted on the left chart of the ICF plots. It maps the channel's amplitude into the "Electrode Activation Level" expressed in the unit that is presented to the clinician in the manufacturer's fitting software. Strategy-specific features (e.g., current steering (Koch 2007)) are not considered in the ICF plots. There is an option to convert the manufacturer-specific level to its general charge equivalent, expressed in nanoCoulomb per pulse phase.

Disclaimer

Neither Otoconsult nor the CI manufacturers can be held legally responsible for any inconsistencies between the graphical representations provided on this web site and the technical CI documentation provided by the CI manufacturers. The use of ICCI and interpretation of the ICF plots is intended as an assistive tool for the competent clinical CI-programmer who remains fully and solely responsible for their use in the clinic.
Otoconsult provides the "ICCI" web site as a demonstrational service only. Otoconsult is not responsible for, and expressly disclaims all liability for, damages of any kind arising out of use, reference to, or reliance on any information contained within the site. While the information contained within the site is periodically updated, no guarantee is given that the information provided in this Web site is correct, complete, and up-to-date.