J Am Acad Audiol 2012; 23(09): 733-750
DOI: 10.3766/jaaa.23.9.7
Articles
American Academy of Audiology. All rights reserved. (2012) American Academy of Audiology

Using Average Correction Factors to Improve the Estimated Sound Pressure Level Near the Tympanic Membrane

Karrie LaRae Recker
,
Tao Zhang
,
Weili Lin
Further Information

Publication History

Publication Date:
06 August 2020 (online)

Background: Sound pressure-based real ear measurements are considered best practice for ensuring audibility among individuals fitting hearing aids. The accuracy of current methods is generally considered clinically acceptable for frequencies up to about 4 kHz. Recent interest in the potential benefits of higher frequencies has brought about a need for an improved, and clinically feasible, method of ensuring audibility for higher frequencies.

Purpose: To determine whether (and the extent to which) average correction factors could be used to improve the estimated high-frequency sound pressure level (SPL) near the tympanic membrane (TM).

Research Design: For each participant, real ear measurements were made along the ear canal, at 2–16 mm from the TM, in 2-mm increments. Custom in-ear monitors were used to present a stimulus with frequency components up to 16 kHz.

Study Sample: Twenty adults with normal middle-ear function participated in this study.

Intervention: Two methods of creating and implementing correction factors were tested.

Data Collection and Analysis: For Method 1, correction factors were generated by normalizing all of the measured responses along the ear canal to the 2-mm response. From each normalized response, the frequency of the pressure minimum was determined. This frequency was used to estimate the distance to the TM, based on the ¼ wavelength of that frequency. All of the normalized responses with similar estimated distances to the TM were grouped and averaged. The inverse of these responses served as correction factors. To apply the correction factors, the only required information was the frequency of the pressure minimum. Method 2 attempted to, at least partially, account for individual differences in TM impedance, by taking into consideration the frequency and the width of the pressure minimum. Because of the strong correlation between a pressure minimum's width and depth, this method effectively resulted in a group of average normalized responses with different pressure-minimum depths. The inverse of these responses served as correction factors. To apply the correction factors, it was necessary to know both the frequency and the width of the pressure minimum. For both methods, the correction factors were generated using measurements from one group of ten individuals and verified using measurements from a second group of ten individuals.

Results: Applying the correction factors resulted in significant improvements in the estimated SPL near the TM for both methods. Method 2 had the best accuracy. For frequencies up to 10 kHz, 95% of measurements had <8 dB of error, which is comparable to the accuracy of real ear measurement methods that are currently used clinically below 4 kHz.

Conclusions: Average correction factors can be successfully applied to measurements made along the ear canals of otologically healthy adults, to improve the accuracy of the estimated SPL near the TM in the high frequencies. Further testing is necessary to determine whether these correction factors are appropriate for pediatrics or individuals with conductive hearing losses.