Do "optimal" conditions improve distortion product otoacoustic emission test performance?

Benjamin J. Kirby, Judy G. Kopun, Hongyang Tan, Stephen T Neely, Michael P Gorga

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

OBJECTIVES: To determine whether an "optimal" distortion product otoacoustic emission (DPOAE) protocol that (1) used optimal stimulus levels and primary-frequency ratios for each f2, (2) simultaneously measured 2f2 - f1 and 2f1 - f2 distortion products, (3) controlled source contribution, (4) implemented improved calibration techniques, (5) accounted for the influence of middle ear reflectance, and (6) applied multivariate analyses to DPOAE data results in improved accuracy in differentiating between normal-hearing and hearing-impaired ears, compared with a standard clinical protocol. DESIGN: Data were collected for f2 frequencies ranging from 0.75 to 8 kHz in 28 normal-hearing and 78 hearing-impaired subjects. The protocol included a control condition incorporating standard stimulus levels and primary-frequency ratios calibrated with a standard SPL method and three experimental conditions using optimized stimuli calibrated with an alternative forward pressure level method. The experimental conditions differed with respect to the level of the reflection-source suppressor tone and included conditions referred to as the null suppressor (i.e., no suppressor tone presented), low-level suppressor (i.e., suppressor tone presented at 58 dB SPL), and high-level suppressor (i.e., suppressor tone presented at 68 dB SPL) conditions. The area under receiver operating characteristic (AROC) curves and sensitivities for fixed specificities (and vice versa) were estimated to evaluate test performance in each condition. RESULTS: AROC analyses indicated (1) improved test performance in all conditions using multivariate analyses, (2) improved performance in the null suppressor and low suppressor experimental conditions compared with the control condition, and (3) poorer performance below 4 kHz with the high-level suppressor. As expected from AROC, sensitivities for fixed specificities and specificities for fixed sensitivities were highest for the null suppressor and low suppressor conditions and lowest for standard clinical procedures. The influence of 2f2 - f1 and reflectance on test performance were negligible. CONCLUSIONS: Predictions of auditory status based on DPOAE measurements in clinical protocols may be improved by the inclusion of (1) optimized stimuli, (2) alternative calibration techniques, (3) low-level suppressors, and (4) multivariate analyses.

Original languageEnglish (US)
Pages (from-to)230-237
Number of pages8
JournalEar and hearing
Volume32
Issue number2
DOIs
StatePublished - Mar 1 2011

Fingerprint

Hearing
ROC Curve
Multivariate Analysis
Clinical Protocols
Sensitivity and Specificity
Calibration
Middle Ear
Ear
Pressure

ASJC Scopus subject areas

  • Otorhinolaryngology
  • Speech and Hearing

Cite this

Do "optimal" conditions improve distortion product otoacoustic emission test performance? / Kirby, Benjamin J.; Kopun, Judy G.; Tan, Hongyang; Neely, Stephen T; Gorga, Michael P.

In: Ear and hearing, Vol. 32, No. 2, 01.03.2011, p. 230-237.

Research output: Contribution to journalArticle

Kirby, Benjamin J. ; Kopun, Judy G. ; Tan, Hongyang ; Neely, Stephen T ; Gorga, Michael P. / Do "optimal" conditions improve distortion product otoacoustic emission test performance?. In: Ear and hearing. 2011 ; Vol. 32, No. 2. pp. 230-237.
@article{23dbdf9f14424b55bed759cf5d6fa8a1,
title = "Do {"}optimal{"} conditions improve distortion product otoacoustic emission test performance?",
abstract = "OBJECTIVES: To determine whether an {"}optimal{"} distortion product otoacoustic emission (DPOAE) protocol that (1) used optimal stimulus levels and primary-frequency ratios for each f2, (2) simultaneously measured 2f2 - f1 and 2f1 - f2 distortion products, (3) controlled source contribution, (4) implemented improved calibration techniques, (5) accounted for the influence of middle ear reflectance, and (6) applied multivariate analyses to DPOAE data results in improved accuracy in differentiating between normal-hearing and hearing-impaired ears, compared with a standard clinical protocol. DESIGN: Data were collected for f2 frequencies ranging from 0.75 to 8 kHz in 28 normal-hearing and 78 hearing-impaired subjects. The protocol included a control condition incorporating standard stimulus levels and primary-frequency ratios calibrated with a standard SPL method and three experimental conditions using optimized stimuli calibrated with an alternative forward pressure level method. The experimental conditions differed with respect to the level of the reflection-source suppressor tone and included conditions referred to as the null suppressor (i.e., no suppressor tone presented), low-level suppressor (i.e., suppressor tone presented at 58 dB SPL), and high-level suppressor (i.e., suppressor tone presented at 68 dB SPL) conditions. The area under receiver operating characteristic (AROC) curves and sensitivities for fixed specificities (and vice versa) were estimated to evaluate test performance in each condition. RESULTS: AROC analyses indicated (1) improved test performance in all conditions using multivariate analyses, (2) improved performance in the null suppressor and low suppressor experimental conditions compared with the control condition, and (3) poorer performance below 4 kHz with the high-level suppressor. As expected from AROC, sensitivities for fixed specificities and specificities for fixed sensitivities were highest for the null suppressor and low suppressor conditions and lowest for standard clinical procedures. The influence of 2f2 - f1 and reflectance on test performance were negligible. CONCLUSIONS: Predictions of auditory status based on DPOAE measurements in clinical protocols may be improved by the inclusion of (1) optimized stimuli, (2) alternative calibration techniques, (3) low-level suppressors, and (4) multivariate analyses.",
author = "Kirby, {Benjamin J.} and Kopun, {Judy G.} and Hongyang Tan and Neely, {Stephen T} and Gorga, {Michael P}",
year = "2011",
month = "3",
day = "1",
doi = "10.1097/AUD.0b013e3181fa5da2",
language = "English (US)",
volume = "32",
pages = "230--237",
journal = "Ear and Hearing",
issn = "0196-0202",
publisher = "Lippincott Williams and Wilkins",
number = "2",

}

TY - JOUR

T1 - Do "optimal" conditions improve distortion product otoacoustic emission test performance?

AU - Kirby, Benjamin J.

AU - Kopun, Judy G.

AU - Tan, Hongyang

AU - Neely, Stephen T

AU - Gorga, Michael P

PY - 2011/3/1

Y1 - 2011/3/1

N2 - OBJECTIVES: To determine whether an "optimal" distortion product otoacoustic emission (DPOAE) protocol that (1) used optimal stimulus levels and primary-frequency ratios for each f2, (2) simultaneously measured 2f2 - f1 and 2f1 - f2 distortion products, (3) controlled source contribution, (4) implemented improved calibration techniques, (5) accounted for the influence of middle ear reflectance, and (6) applied multivariate analyses to DPOAE data results in improved accuracy in differentiating between normal-hearing and hearing-impaired ears, compared with a standard clinical protocol. DESIGN: Data were collected for f2 frequencies ranging from 0.75 to 8 kHz in 28 normal-hearing and 78 hearing-impaired subjects. The protocol included a control condition incorporating standard stimulus levels and primary-frequency ratios calibrated with a standard SPL method and three experimental conditions using optimized stimuli calibrated with an alternative forward pressure level method. The experimental conditions differed with respect to the level of the reflection-source suppressor tone and included conditions referred to as the null suppressor (i.e., no suppressor tone presented), low-level suppressor (i.e., suppressor tone presented at 58 dB SPL), and high-level suppressor (i.e., suppressor tone presented at 68 dB SPL) conditions. The area under receiver operating characteristic (AROC) curves and sensitivities for fixed specificities (and vice versa) were estimated to evaluate test performance in each condition. RESULTS: AROC analyses indicated (1) improved test performance in all conditions using multivariate analyses, (2) improved performance in the null suppressor and low suppressor experimental conditions compared with the control condition, and (3) poorer performance below 4 kHz with the high-level suppressor. As expected from AROC, sensitivities for fixed specificities and specificities for fixed sensitivities were highest for the null suppressor and low suppressor conditions and lowest for standard clinical procedures. The influence of 2f2 - f1 and reflectance on test performance were negligible. CONCLUSIONS: Predictions of auditory status based on DPOAE measurements in clinical protocols may be improved by the inclusion of (1) optimized stimuli, (2) alternative calibration techniques, (3) low-level suppressors, and (4) multivariate analyses.

AB - OBJECTIVES: To determine whether an "optimal" distortion product otoacoustic emission (DPOAE) protocol that (1) used optimal stimulus levels and primary-frequency ratios for each f2, (2) simultaneously measured 2f2 - f1 and 2f1 - f2 distortion products, (3) controlled source contribution, (4) implemented improved calibration techniques, (5) accounted for the influence of middle ear reflectance, and (6) applied multivariate analyses to DPOAE data results in improved accuracy in differentiating between normal-hearing and hearing-impaired ears, compared with a standard clinical protocol. DESIGN: Data were collected for f2 frequencies ranging from 0.75 to 8 kHz in 28 normal-hearing and 78 hearing-impaired subjects. The protocol included a control condition incorporating standard stimulus levels and primary-frequency ratios calibrated with a standard SPL method and three experimental conditions using optimized stimuli calibrated with an alternative forward pressure level method. The experimental conditions differed with respect to the level of the reflection-source suppressor tone and included conditions referred to as the null suppressor (i.e., no suppressor tone presented), low-level suppressor (i.e., suppressor tone presented at 58 dB SPL), and high-level suppressor (i.e., suppressor tone presented at 68 dB SPL) conditions. The area under receiver operating characteristic (AROC) curves and sensitivities for fixed specificities (and vice versa) were estimated to evaluate test performance in each condition. RESULTS: AROC analyses indicated (1) improved test performance in all conditions using multivariate analyses, (2) improved performance in the null suppressor and low suppressor experimental conditions compared with the control condition, and (3) poorer performance below 4 kHz with the high-level suppressor. As expected from AROC, sensitivities for fixed specificities and specificities for fixed sensitivities were highest for the null suppressor and low suppressor conditions and lowest for standard clinical procedures. The influence of 2f2 - f1 and reflectance on test performance were negligible. CONCLUSIONS: Predictions of auditory status based on DPOAE measurements in clinical protocols may be improved by the inclusion of (1) optimized stimuli, (2) alternative calibration techniques, (3) low-level suppressors, and (4) multivariate analyses.

UR - http://www.scopus.com/inward/record.url?scp=79952315402&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952315402&partnerID=8YFLogxK

U2 - 10.1097/AUD.0b013e3181fa5da2

DO - 10.1097/AUD.0b013e3181fa5da2

M3 - Article

VL - 32

SP - 230

EP - 237

JO - Ear and Hearing

JF - Ear and Hearing

SN - 0196-0202

IS - 2

ER -