Tous droits réservés © NeurOreille (loi sur la propriété intellectuelle 85-660 du 3 juillet 1985). Ce produit ne peut être copié ou utilisé dans un but lucratif.
The cochlea deconstructs a sound in the same manner as a band-pass filter bank. This section describes how a cochlear implant speech processor transforms auditory acoustic information into an electric signal delivered to each intracochlear electrode, taking into account the tonotopic organisation of the cochlea.
The acoustic signal picked up by the microphone (1) is amplified (2) and sent to a filter bank (3), which breaks it down into n frequency bands (n being the number of intracochlear electrodes).
The output from the filter is sent to an envelope detector (4), which constantly determines the energy of the signal for each frequency band. A dynamic range modulator (5) is used to adapt every output to the patient perceptual thresholds. Finally, the processor (6) generates electrical stimuli proportional to the energy of the input signal. These are sent to the implanted electrodes (7), following the cochlear tonotopic organization.
To avoid interaction between neighboring electrodes, a sequential strategy is applied. A common strategy is called n-of- m, meaning that at each cycle, the electrodes that are activated correspond to the frequency bands that contain the most energy (n) among all available frequency bands (m).
Sudden electrical stimulation of nerve fibers is more efficient than gradual increases in electrical energy, which is why rectangular pulses are generally used. To avoid electrolytic polarization, which can be toxic to the nerve and surrounding tissues, biphasic stimuli are applied. Stimulation level is adjusted by modifying the two parameters that account for the amount of the delivered energy: voltage and duration.