Cochlear implants: Modeling electrophysiological responses

Cochlear implants (CI) are implantable devices to rehabilitate hearing for people with severe to profound hearing loss. Over recent years, introduction of new designs have not lead to significant improvements in outcomes. Tests to evaluate new sound coding strategies and recording methods are time-consuming, and yield results that are difficult to interpret due to a relatively low number of patients and a large heterogeneity in etiology. As an alternative approach to test new designs, a computational model can be used. In this thesis two computational models are presented and validated for simulation of peripheral responses in cochlear implant users. One model simulates spike times in response to electrical pulse trains. In this model, a three-dimensional volume conduction of the cochlea and an active nerve model were extended with stochasticity, adaptation, and accommodation. It can be used to simulate single fiber action potentials (SFAP) and evoked compound action potentials (ECAP). The second model simulates electrocochleography (eCochG) potentials as recorded intracochlearly. Both models can be used to test developments in sound coding and new recording methods. Furthermore, it can be tested how differences in the individuals’ auditory peripheral system contributes to differences in objective recordings obtained from the individuals. This research was funded by: Advanced Bionics Corporation and Stichting Het Heinsius-Houbolt Fonds. This thesis was financially supported by: Advanced Bionics, Cochlear, MED-EL, Oticon, Pedagogische Praktijk, Chipsoft, Amplifon – Beter Horen .LUMC / Geneeskund