[Sound and velocity DPOAEs : Technology, methodology and perspectives]

Abstract

Recent publications show that DPOAE measurements can generate a more accurate diagnosis, if (1) their fine structure is suppressed, and (2) if the calibration of the sound field is improved. Reduction of the fine structure is particularly important in the frequency range below 4 kHz in subjects with intact cochlear amplifier and can reduce the standard deviation of threshold estimations based on DPOAE-input/output functions from 11 dB to 6 dB. Improving the sound-field calibration has most impact in the frequency range above 4 kHz. Threshold estimations based on laserinterferometrically measured DPOAE input-output functions where the sound field was calibrated close to the tympanic membrane have been shown to reduce the standard deviation down to 8.6 dB in humans and 6.5 dB in guinea pigs. Compared with conventional DPOAE measures, such as amplitude or signal-to-noise ratio, threshold estimation based on DPOAE-I/O functions has the advantage that its slope provides additional information about the middle-ear; however, its specificity is limited. In the future, combined methods such as acoustic reflectance or laser vibrometry on the umbo promise a reliable assessment of the middle-ear contribution to DPOAE.