Evaluation of a Fast Method to Measure High-Frequency Audiometry Based on Bayesian Learning.
Chiara Casolani, Ali Borhan-Azad, Rikke Skovhøj Sørensen and 2 others
PMID 38192152WHAT IT FOUND
A Bayesian algorithm measured high-frequency hearing thresholds in about 4 minutes, compared to 13 minutes for a forced-choice test.
It was reliable and similar to standard manual audiometry, though it reported slightly lower thresholds. This speed could make high-frequency screening practical in clinics.
Key findings
01The Bayesian test averaged 3 minutes and 40 seconds, while the forced-choice test averaged 12 minutes and 53 seconds.
02The Bayesian method showed high test-retest reliability, with intraclass correlation coefficients above 0.90 for most groups.
03Automated tests produced systematically lower threshold estimates than standard manual audiometry, a bias that can be compensated for.
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The study population had only mild hearing impairment, so results may not generalize to those with severe loss. Headphone placement was not standardized or monitored, which can affect high-frequency thresholds. Participants with tinnitus were tested with warble tones, while others used pure tones, introducing a stimulus difference. The standard manual audiometry test duration was not recorded, preventing a direct time comparison with the automated methods. The study did not compare these methods against objective measures like hearing aid fitting outcomes.
Declared interests
The authors declared no conflicts of interest. The study was funded by William Demant Fonden and the TIN-ACT EU innovative training network.
The easy way to misread this
Do not assume the Bayesian test is interchangeable with standard audiometry without adjustment. It consistently reports lower thresholds than the manual method, so using its raw values without correcting for this bias could lead to underestimating hearing loss.