Signal processing approaches on otoacoustic emissions

The recent achievement on the measurement of otoacoustic emissions (OAEs) is based on a novel technical development of digital signal processing. OAEs measured in the external ear canal are normal by-products of the active process in hearing, which was discovered by Kemp (1978). Outer hair cells (OHCs) are thought to be the active source in the generation of this energy. Signal processing methods play a crucial role in the detection of OAEs in noise and artifacts, and in the extraction of information from OAE recordings.

The present thesis is focused on the signal processing methods used in the recording, data representation, and information extraction of OAEs:

(1) A time-frequency method for analysis of transient evoked OAEs (TEOAEs) via smoothed pseudo Wigner distribution has been developed. TEOAEs can be transformed into the time-frequency plane to give a three-dimensional pattern. The analysis of shape and localization of TEOAE pattern and the comparison of pattern differences establish a method to extract more information from TEOAEs.

(2) An optimal recording protocol based on time-frequency analysis of TEOAEs has been proposed for neonatal hearing screening. A better signal-to-noise ratio (SNR) and a lower noise level of TEOAEs have been achieved by shortening the recording window and by using a linear recording protocol. The method has been applied in three audiological clinics in Europe. Time-frequency analysis of TEOAEs indicates a significantly reduced energy in the mid to high frequency bands for subjects with sensorineural hearing loss (SNHL) compared to normal-hearing subjects.

(3) TEOAEs, spontaneous OAEs (SOAEs) and distortion-product OAEs (DPOAEs) are related. The contribution from synchronized SOAEs to TEOAEs was demonstrated. The female and right ear advantages on OAEs were observed.

(4) Spectral estimation of SOAEs was performed by an average periodogram, a reduced variance estimate, and a model based high-order autoregressive (AR) estimate. Different spectral estimation methods can give more information on the spectral pattern of SOAEs.

(5) Active cochlear nonlinearity was estimated by multi-component DPOAEs and by introducing generating models of DPOAEs. The input-output function of the active cochlear nonlinearity was calculated from the multi-component DPOAEs. The results show that the generating mechanism of DPOAEs is dependent on stimulus level.

(6) The "bounce" phenomenon on basilar membrane nonlinearity was observed after exposure to a loud, but not traumatic low-frequency tone. This may give objective information on an individual's ability to recover from a temporal threshold shift (TTS).

In summary, the importance of these results relies mainly on the refinements of the measurement tools created, which can be used to investigate the function of the inner ear, especially the outer hair cells (OHCs).

List of scientific papers

I. Cheng J (1995). "Time-frequency analysis of transient evoked otoacoustic emissions via smoothed pseudo Wigner distribution" Scand Audiol 24(2): 91-96
https://pubmed.ncbi.nlm.nih.gov/95389095

II. Hatzopoulos S, Cheng J, Grzanka A, Morlet T, Martini A (2000). "Optimization of TEOAE recording protocols: a linear protocol derived from parameters of a time-frequency analysis: a pilot study on neonatal subjects" Scand Audiol 29(1): 21-27
https://pubmed.ncbi.nlm.nih.gov/20181601

III. Hatzopoulos S, Cheng J, Grzanka A, Martini A (2000). "Time-frequency analyses of TEOAE recordings from normals and SNHL patients" Audiology 39(1): 1-12
https://pubmed.ncbi.nlm.nih.gov/20210623

IV. Cheng, J (1998). "Otoacoustic emissions: Measurement, data and interrelations" Acustica - acta acustica 84: 320-328

V. Cheng, J (1998). "Spectral estimation of spontaneous otoacoustic emissions" Acustica - acta acustica 84: 712-719

VI. Cheng, J (1999). "Estimation of active cochlear nonlinearity by multi-component distortion-product otoacoustic emissions" Acustica - acta acustica 85: 721-727

VII. Cheng, J (2000). "Quantifying basilar membrane nonlinearity and the "bounce" phenomenon on the nonlinearity estimated by multi-component DPOAEs" (Manuscript)