Reduction of low-frequency noise in click-evoked otoacoustic emissions. An optimal filtering approach

Noise reduction is a problem of practical concern in the measurement of click-evoked otoacoustic emissions. Instrumentation and environmental noise are suppressed by synchronous averaging and rejection-level techniques that are technically used in most recording systems. Patient noise (such as swallowing, breathing, heart beat) is often of low frequency and can be reduced by digital filtering. A high-pass filtering approach is proposed to reduce low-frequency noise. Cross-correlation was used to determine the cut-off frequency that maximizes the S/N ratio. Click-evolved emissions from normal and hearing impaired adults have been used as test materials. Preliminary results show a remarkable improvement in the S/N ratio of the recorded trials