Project name: Validation of Smartphone-based Hearing Test Results
Investigator(s): Richard L Neitzel (PI), Glenn Green (Co-I)
Team: Xin Zhang, Lauren M Smith, Joyce Daniels, Zachary Graham
Source of funding: Apple Inc
Dates: March 2024 – present
Description:
Background: Evaluations of self-administered hearing tests using personal listening devices are limited. This experimental and observational study, conducted among participants with normal or impaired hearing, aims to examine the differences between app-based and audiometer-based (reference) hearing test results.
Methods: Each participant will complete a series of pure-tone audiometry (PTA) and speech-in-noise (SIN) tests using a designated mobile phone app in an audio booth, utilizing six types of popular personal listening devices. The reference hearing threshold will be determined by averaging four thresholds (4 PTA, i.e., at 500, 1k, 2k, and 4k Hz) evaluated using the GSI Pello with AMTAS software. The reference SIN score is the Speech-to-Noise Ratio (SNR) obtained from QuickSIN using the Tillman-Olsen formula.
The median of the absolute differences (MAD) between the app-based and paired audiometer-based 4PTA thresholds will be used to evaluate the accuracy of the app-based tests. The MAD between repeated app-based tests using the same listening device will be used to assess performance consistency. The MADs will be bootstrapped to obtain estimated confidence intervals (CIs) for each listening device at each frequency. A confusion matrix will be used to report the test’s ability to accurately classify degrees of hearing loss for early detection. Speech-in-noise test results will be compared using the percentage of words correctly repeated.
Objectives or Specific Aims:
- To assess the sensitivity, specificity and predictive probability of the mobile app-based pure tone audiometry in classifying hearing loss
- To assess the accuracy and precision of the mobile app-based pure tone audiometry in measuring the hearing threshold
- To assess the sensitivity, specificity and predictive probability of the mobile app-based speech-in-noise test in classifying speech recognition ability
- To assess the accuracy and precision of the mobile app-based speech-in-noise test in measuring the speech to noise ratio
- To assess the list equivalency of the speech-in-noise test materials
- To obtain a set of adjustment factors for pure tone audiometry at each frequency to be applied in studies that utilize such hearing tests
Exposure Research Lab 