COCHLEAR IMPLANT SIMULATION       version 2.0              

Main page

What is not “Cochlear Implant Simulation”?

The aim of the software “Cochlear Implant Simulation” is to simulate the auditory perception through the cochlear implant. This simulation is based on a series of considerations and models. We could affirm that the signals synthesized by this software represent appropriately the perception through the cochlear implant if the considered models represent appropriately the process of perception and the selected parameters are set to the appropriate values, according to the situation to be modeled. The results of the validation tests guarantee the adequation of the synthesized signals as an approximation to the perception through a cochlear implant. However, it is almost impossible to establish in an unequivocal and definitive way that a synthesized signal represents exactly the perception through the cochlear implant. In that sense, the results of this simulation should be considered with some prudence.

The aspects not considered in the simulation that could influence the perception make that the synthesized signals do not represent exactly the perception through the cochlear implant. Among others aspects not modeled in the developed software, the next ones can be enumerated:

• The fitting map in the processor. In the simulation, it is assumed that the fitting map is perfectly adapted to the requirements of the implanted patient, i.e., there is not loss of information due to an improper adjustment of the processor.
• The resolution in intensity. In the simulation software, the limitation of the resolution in intensity associated to the neural ends is not modeled, and therefore the degradation in the quality associated to a limited resolution in intensity is not represented. This would be particularly important in the case of patients with a low index of surviving neurons in the cochlear nerve.
• Variation of the physiological parameters along the cochlea. An adequate and accurate model of the interaction electrode - auditory nerve should consider that the channel interaction coefficient could be different for the different cochlear partitions. It also should consider that the refractory period of the neurons, the percentage of surviving neurons, the resolution in intensity and other parameters could present variations along the cochlea.
• Modeling abnormal cochleas, like ossified or malformed cochleas. In the simulation software, we have assumed that the implant is allocated into a permeable cochlea and that the electrodes are allocated being the first one of the electrode carrier in the most apical position, and the last one in the most basal position, which allows the estimation of the characteristic frequency associated to each electrode according to the place theory. The allocation of the electrodes in the case of ossified cochleas (when the trajectory of the implant does not follow the scala tympani or when the a cochlear implant with a split electrode carrier is used) or in the case of malformed cochleas (like common cavities) is difficult to be modeled and has not been considered in the developed software.

Other important aspect to be considered is the fact that in this simulation we have estimated what would be the pattern of activity in the neural ends of the auditory nerve, and this pattern has been considered the starting point for the synthesis of the audio signal. However, in this software we did not make any consideration about the propagation of the neural activity through the auditory pathway or what is the signal processing affecting the pattern of action potentials when they are transmitted through the auditory brainstem. Modeling this aspect would be extremely complex due to the limited knowledge about the role of the auditory brainstem in hearing perception. It could be assumed that in the case of patients with hearing experience, the signal processing performed in the auditory brainstem would be “normal” (similar to that for normal-hearing subjects) but in the case of absence of hearing experience, the lack of maturation of the auditory pathway would cause an additional loss of information, not considered in the simulation software. The development of the hearing skill by the implanted patients (or the signal processing, or processing of information at the cortical level) would be also difficult to be modeled (and has not been modeled) and also affects the way in which the sound is perceived through the cochlear implant.

All these aspects must be taken into account when one affirms that the synthesized signals represents, in some way, the perception of sound through the cochlear implant. In future revisions of the software “Cochlear Implant Simulation” we plan to include some of the above described effects and factors.