Factors to consider when measuring DPOAE’s

Audiologists who use distortion product otoacoustic emission (DPOAE) measures in the clinical setting need to be aware of two possible measurement variables that may introduce errors into their results.  They are the ambient noise level in their testing environment and the frequency specific averaging time used during the measurement.  In many clinical environments, DPOAE measures are not conducted in a sound treated booth, and ambient noise level may interfere with the emission measurement.    

The measure that is of interest when examining ambient room levels is, the D:N (DPOAE level to noise ratio).  Lee et al. 1999 conducted a study to measure the D:N ratio as well as the effects of averaging times used during measurements.  A DPOAE can be said to be present if the D:N ratio is above 0.

When measuring DPOAE’s, adequate averaging times are required.  Lee et al 1999 used four different averaging times ranging from 1.3 to 10.4 seconds.  In order to be certain that acoustic energy is due to the cochlear emission rather than to non-linear effects of the equipment, Lee et al 1999 conducted measurements in deaf ears.       

A base line condition was established where the ambient noise level of 25 dBA was maintained.  Lee et al 1999 indicated that the maximum permissible ambient noise level (MPANL) in their study, for a given condition was estimated as the highest level of ambient noise for which the D:N was not significantly different than the D:N of the baseline condition, and where 95% of normal ears showed detectable emissions.

According to Lee et al 1999, ambient noise greater than 40-50 dBA could have considerable effect on the D:N for low frequencies (0.7 – 1 kHz).  This however is dependent on the measurement equipment being used.  In their study, an Otodynamics IL092 Distortion Product Analyzer was used.  For middle frequencies (1.5-2 kHz), a nose level of 45-55 dBA is tolerable).  At test frequencies above 2 kHz, ambient noise levels above 60 or 65 dBA is not likely to affect the D:N ratio.  Audiologists will find this quite logical when the physical properties of noise is explained.  Noise, or in this case, pink noise will have most energy concentrated in the lower frequencies, and this is the region which most adversely affects the measurement of the DPOAEs.   

For Audiologists, what is interesting to know from the results of the Lee et al 1999 study is that longer averaging times were needed for lower frequencies in order to obtain detectable DPOAEs.  The graphs below illustrate these results nicely.  Audiologists need to be aware that when measuring DPOAEs across different frequencies, longer averaging times may be required for lower frequencies for measured emissions to be detectable.  Not doing so may result in the occurrence of a DPOAE being lost in the noise floor.     

click to enlarge

Audiologists should be aware that even in extremely noisy environments, a DPOAE response can be obtained with averaging up to 10.4 seconds at frequencies above 1 kHz.  The drawback in testing in a noisy environment is the greater time required to complete the testing. 

On a final note, I would like to indicate that although the study conducted by Lee et al 1999 may be mostly generalized across most testing situations, (that is the noisier the environment, the longer the averaging time required.etc), it is specific to the equipment being used.  During the third annual Canadian Academy of Audiology Conference in Toronto, I had the opportunity to play with a GSI Automated OAE and was able to obtain an OAE in a noisy conference showroom.   

Note:  This site is best viewed using IE 4.X or higher under 800x600 resolution