Artikel
Finite element modeling of neuroelectronic interfaces as development tool for neuroprostheses
Finite Elemente-Modellierung neuroelektronischer Schnittstellen als Werkzeug zur Optimierung von Neuroprothesen
Suche in Medline nach
Autoren
Veröffentlicht: | 11. April 2007 |
---|
Gliederung
Text
Objective: Neuroprosthetics is a promising modern field of restorative surgery. The functional recovery of hearing and visual functions as well as the voluntary control of artificial limbs are currently focussed applications. Due to miniaturisation and high performance computer systems limitations by signal processing became less important. The bottleneck according to signal transmission is caused by the interface between electronic and nervous system. The signal transmission is limited by the number and conductive properties of the used surface or penetrating electrodes. Over time scaring and chronic damage of the nervous tissue compromise the interface. Our aim was to develop an environment for simulating and testing different influencing variables beginning with the biomechanical aspects.
Methods: Based on MR images (3D-MPR, T2w) of a brainstem a 3D-finite element (FE) model was generated with automated routines developed in a prior research project. The brain tissue material properties were assigned corresponding to the MRI-intensity values and data of the literature. Different geometrically designed penetrating and surface electrodes were integrated into the model simulating auditory brain stem implants. Parameter studies were performed with varying material properties of the electrodes. Simulated stress and strain values were evaluated.
Results: The material properties of the generated FE-model are nonlinear implemented for the brain tissue. Surface- and penetrating electrodes with varying material properties were successfully integrated and tested. High tissue stresses were detected at the tip of penetrating electrodes with straight pin-design. These stress-peaks could be reduced both, by electrodes with clubbed tips and discontinuous material properties along the pin-axis. Surface electrodes with small diameter showed minimal stress values.
Conclusions: The developed FE-environment was successfully tested to evaluate stress-distributions for different geometric and material properties of brain stem electrodes. Further developments of this software may help to optimise neuroelectronic interfaces. New Electrode-designs and applications can be tested and preselected to avoid unnecessary experimental in vivo studies, not only contributing to a cost reduction. New software applications for the evaluation of electric and thermal effects at the neuroelectronic interface are in the development stage and offer promising new application fields.