An interactive evolutionary algorithm for cochlear implant fitting

Abstract

In a previous SAC-COMPAHEC paper[1], a method was presented using an interactive evolutionary algorithm for cochlear implants fitting.The method has recently been put to test, with very unexpected and encouraging results: in a few words, it seems that the algorithm is capable to obtain much better results than an expert practitioner in many cases.The solutions proposed by the algorithm are counter-intuitive, yet they improve speech recognition drastically. If these preliminary results are confirmed by many more cases, it could mean that experts have been deterministically tuning cochlear implants the wrong way for many years.However, it seems that the very good results obtained by the algorithm depend a lot on the acoustic environment in which the fitting is performed. A broader fitting scheme has therefore been implemented that should overcome this problem, by allowing the patient to sample typical background noises for which the prosthesis should be specifically tuned.In the future, a piece of software will be added to the cochlear implant signal processor that will automatically choose the best setting depending on the kind of sound environment picked up by the microphone.

Interactive evolution for cochlear implants fitting

“…thesis of the Université de Bourgogne. These results have already been presented in several papers [6,7,8].…”

Simulating listener errors in using genetic algorithms for perceptual optimization

“…Previous studies had proposed optimization algorithms for customizing hearing aids (Durant et al, 2004;Kuk and Pape, 1992;Neuman et al, 1987;Preminger et al, 2000;Stelmachowicz et al, 1994) or cochlear implants (Bourgeois-République et al, 2005;Wakefield et al, 2005) with real patients. Even though the simulation results of the present study would be applicable specifically to the GA implementation reported by Başkent et al (2007b), similar simulations could be used to characterize the effects of differing user skills on how well any perceptual optimization method might work for general population.…”

“…The modified simplex algorithm was proposed for fitting gain in hearing aids (Kuk and Pape, 1992;Neuman et al, 1987;Preminger et al, 2000;Stelmachowicz et al, 1994). Genetic algorithms (GAs) were suggested for fitting features related to hearing aids (Durant et al, 2004) or cochlear implants (Bourgeois-République et al, 2005;Wakefield et al, 2005). In such perceptual optimization, candidate parameter sets are first evaluated by a listener and then modified according to listener's preferences following the rules of the particular method used.…”

Eye-tracking evolutionary algorithm to minimize user fatigue in IEC applied to interactive one-max problem

“…We can also mention work in combining human interactions with an artificial ant, applied to non-photorealistic rendering [15]. Another use of IEC involves a human patient using a PDA on which an IEC is launched to define best parameter values for cochlear implants [2]. First results show that patients using PDAs obtain a better parameterization than previously through lengthy interaction with a doctor.…”