Evaluating the noise in electrically evoked compound action potential measurements in cochlear implants

Area of Science:

• Auditory Neuroscience
• Biomedical Engineering
• Otolaryngology

Background:

• Electrically evoked compound action potentials (ECAPs) assess auditory nerve function in cochlear implant (CI) users.
• ECAP detection is often subjective, especially at low stimulation levels or for spread of excitation.
• Objective criteria are needed to improve ECAP measurement reliability.

Purpose of the Study:

• To evaluate statistical properties of background noise (BN) and residual noise (RN) in ECAP measurements.
• To establish an objective detection criterion for ECAPs.
• To assess the benefit of weighted averaging for ECAP analysis.

Main Methods:

• Applied a noise estimation method used in auditory brainstem response measurements.
• Analyzed approximately 360 ECAP measurements from five human CI users (CII or HiRes90K devices).
• Investigated the impact of stimulation intensity and electrode position on noise levels.

Main Results:

• Background noise (BN) followed a normal distribution.
• Residual noise (RN) decreased with the square root of the number of averages.
• Weighted averaging provided no significant benefit.
• Noise levels were consistent across stimulation intensities and recording electrodes.
• A signal-to-noise ratio (SNR) of 1.7 dB yielded a 1% false positive detection rate.

Conclusions:

• Established an objective detection criterion for ECAPs based on SNR.
• Demonstrated that noise characteristics are stable across various stimulation and recording conditions.
• The findings facilitate more reliable and objective ECAP measurements in CI research and clinical practice.