Finite element study on chochlear implant electrical activity

Waldo Nogueira, Andreas Buechner

Deutsches Hoerzentrum Hannover, Hearing4all, Hannover Medical School


Cochlear Implants (CIs) are medical devices used to restore the sense of hearing in people with profound hearing loss. In a cochlear implant the auditory nerve inside the inner ear or cochlea is stimulated with electrical pulses by means of implanted electrodes. These devices allow in general for good speech communication in non-noisy environments and produce limited music perception. One major issue with these devices is the large variability in performance observed among CI users. One possible reason that might explain, at least partially, the large variability observed, is the difference in the interface created by the electrodes and the auditory nerve for each cochlear implant user. For example there are two unknowns in the auditory nerve interface that might influence performance 1) the exact position of the electrode in the 3D space of the cochlea and, 2) the amount of functional auditory neurons. 

One possible solution to reduce the variability in performance would be to understand the individual electrode-nerve interface and consequently adapt the CI device to the needs of each user.   With this purpose we propose to use a model of the electrically excited cochlea combined with an auditory nerve model to understand the individual differences in electrode position and amount of neural survival in each individual CI user.

Finite element (FE) models of cochlear implants have been developed in the past to model the electrical field in the cochlea [1,2].  In this work a 3D mesh has been created based on single midmodiolar horizontal section of a human cochlea.  The mesh models one and a half turns of the cochlea and is composed by different compartments and membranes (the scala timpany, scala vestibuli, scala media, reissener membrane and basilar membrane).  The electric field in the cochlea is estimated after stimulation with a realistic electrode array inserted along the scala timpany.  Additionally, a model of the auditory nerve fibers is used to predict the amount of neural activity when excited by the estimated electrical field.

Results and Conclusion
The electrical field and neural excitation patterns created by different electrode geometries and positions have been evaluated for different electrode configurations. For a standard electrode insertion, the current model was able to estimate the electrical field patterns produced by tripolar, bipolar and monopolar electrode stimulation. As expected, the model predicts larger excitation patterns for monopolar than for bipolar, and larger excitation patterns for bipolar than for tripolar stimulation.

This work has been funded by the Excellence Cluster Initiative Hearing4all.

[1] Frank Rattay, Richardson Naves Leao, Heidi Felix, “A model of the electrically excited human cochlear neuron. II. Influence of the three-dimensional cochlear structure on neural excitability”, Hearing Research, 153 , 64^79, 2001.

[2] Svante Stadler and Arne Leijon, “Prediction of Speech Recognition in Cochlear Implant Users by Adapting Auditory Models to Psychophysical Data”,  EURASIP Journal on Advances in Signal Processing Volume 2009.