Optimization of Cochlear Implant Fitting in Patients With Functional Contralateral Hearing Using an Evolutionary Algorithm.

360 million people worldwide suffer from disabling hearing loss. The prevalence of hearing impairment, all stages combined, in the French population is 7% (4 million people), of which 9% have severe impairment and 3% have profound or total impairment.

Cochlear implants are indicated in severe to profound deafness in some cases. The principle of the cochlear implant is to directly stimulate the fibres of the auditory nerve via electrodes inserted into the cochlea. It stimulates the auditory nerve and sends electrical impulses to the brain where they are interpreted as sounds.

The steps in hearing rehabilitation are surgical placement of the cochlear implant, activation, and follow-up adjustments.

There is no formal consensus on the exact adjustment procedures during activation or follow-up, but principles are followed depending on the cochlear implant adjustment centers (jack). All centers focus on sound intensity adjustments to achieve the goals of tonal audiometry in open-field silence with cochlear implant alone between 20 and 40 db (30 db most frequently). However, it is common practice to observe that this means of assessment does not really represent the performance of the individual because hearing a sound does not mean that someone will be able to recognize it and interpret it. Some of the patients who achieve these goals have difficulty hearing well in a noisy environment. Speech audiometry in silence and especially in noise would be a better reflection of patient needs.

Bimodal hearing is having a cochlear implant and a contralateral hearing aid. In cochlear implants, having bimodal binaural hearing improves the patients ability to understand speech in silent and noisy contexts. It should thus be considered when a second cochlear implant is not indicated in the contralateral ear.

It has been shown that intelligibility and musical perception are altered when the frequency allocations are different from the manufacturer's default frequency allocations. The possible redundancy between the acoustic and electrical information on the contralateral ear can lead to metallic distortion of the voice, which is perceived as less natural. This is due to a different stimulation of the cochlear tonotopic zones concerning conversational frequencies, between the implanted side and the device side. However, the frequency distribution is not subject to adjustment in current practice. A modification of the frequency distribution is possible on cochlear implants, which could improve the intelligibility and comfort of the implanted patient.

This working hypothesis will be studied, and a simple protocol for frequency reallocation of the cochlear implant will be developed to optimize the daily hearing performance of the implanted patients. An evolutionary algorithm will be used. The search for new adjustment solutions will be carried out within the safety limits imposed by the adjusters (detection and comfort loudness threshold).