Computational Evaluation of Cochlear Implant Surgery Outcomes Accounting for Uncertainty and Parameter Variability

Area of Science:

• Biomedical Engineering
• Computational Neuroscience
• Medical Imaging

Background:

• Cochlear implantation (CI) restores hearing but outcomes vary due to unpredictable surgical factors.
• Accurate computation of current propagation and neural activation is hindered by uncertainties in electrode position and bone electrical properties.

Purpose of the Study:

• To quantify the impact of uncertainties on cochlear implant computational simulations.
• To develop a framework for assessing neural response variability and confidence intervals.

Main Methods:

• Utilized an automated framework for finite element generation and neural response assessment.
• Employed a statistical shape model and Monte Carlo sampling for population variability.
• Applied the Probabilistic Collocation method for patient-specific uncertainty propagation.

Main Results:

• Bone resistivity was identified as the most influential factor on CI outcomes.
• Smaller cochleae with high bone resistivity correlated with poorer outcomes.
• The Probabilistic Collocation method offered a balance between accuracy and computational cost.

Conclusions:

• The developed framework can aid in surgical planning and audiological fitting for cochlear implants.
• Understanding uncertainty propagation is crucial for optimizing CI performance.
• Bone resistivity and cochlear length are key parameters influencing CI success.