Bio-Inspired Cochlear Implant Fitting Through Neural Response Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cochlear implant fitting methods are time-consuming and require significant clinical experience, often lacking patient feedback, especially for children, and existing evaluation tools fail to accurately predict speech intelligibility due to coarse neurogram analysis.
Innovation Solution
A method combining objective measurements like ECAP and ESRT with neural response models to automatically adjust cochlear implant settings, using a parameter adjustment algorithm to match normal hearing neural responses, incorporating subjective feedback and environmental conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional cochlear implant fitting methods are used, then clinical experience and manual adjustment are required, but the fitting process becomes time-consuming and inefficient
Solution Approach 1:
The system enables automatic self-adjustment of cochlear implant parameters through neural response modeling. The processor automatically analyzes neural responses and adjusts stimulation parameters without requiring extensive manual clinical adjustment, allowing the system to optimize itself based on measured neural data and simulated hearing perceptions.
Solution Approach 2:
The patent replaces manual mechanical adjustment processes with computational modeling and simulation. Instead of relying on clinician expertise and trial-and-error adjustment, the system uses neural response models and virtual hearing simulations to automatically determine optimal implant settings, substituting mechanical/manual processes with digital computation.
2Measurement precision
If coarse neurogram analysis is used for evaluation, then the analysis is simpler and faster, but the prediction of speech intelligibility becomes inaccurate
Solution Approach 1:
The system creates virtual copies or simulations of neural responses and hearing perceptions through computational models. By simulating neural processing and hearing outcomes in a virtual environment, the system can accurately predict speech intelligibility without requiring complex physical measurements, using digital twins to represent physiological responses.
Solution Approach 2:
The patent transitions from coarse, low-dimensional neurogram analysis to multi-dimensional neural response modeling. By incorporating multiple neural response parameters, temporal patterns, and spatial distributions into the analysis, the system achieves higher prediction accuracy by moving from a single-dimensional to a multi-dimensional evaluation space.
3Productivity
If automatic adjustment algorithms are implemented, then fitting time is reduced, but the complexity of the system increases
Solution Approach 1:
The system performs preliminary computational work by pre-calculating and storing neural response models and hearing simulations before the actual fitting process. By preparing these models in advance, the system can quickly match and adjust parameters during fitting without performing complex calculations in real-time, reducing overall system complexity while maintaining high efficiency.
Data Source
AI summary
Arrangements are described for fitting an implanted patient and a hearing implant system having an implanted electrode array of electrode contacts. Objective response measurements are performed following delivery of preliminary electrical stimulation signals to the electrode contacts to determine a preliminary fit map that characterizes preliminary patient-specific operating parameters for the hearing implant system. Then an adjusted fit map is produced that characterizes adjusted patient-specific operating parameters for the hearing implant system based on using the preliminary fit map to constrain an implant neural response model to best fit a normal hearing neural response model.