Estimating Hearing Thresholds From Stimulus-Frequency Otoacoustic Emissions.
Qin Gong, Yin Liu, Zewen Peng
PMID 32965182WHAT IT FOUND
SFOAEs could estimate hearing thresholds and identify normal versus impaired hearing with high accuracy, especially at 0.5 to 4 kHz.
Accuracy was poorer for mild and severe losses and at 8 kHz. Testing still took several minutes per frequency.
Key findings
01The neural-network method estimated thresholds more often and more accurately than the linear method across the tested frequencies.
02SFOAEs identified whether hearing was normal or impaired with high accuracy at 0.5 to 4 kHz, and lower accuracy at 8 kHz.
03Threshold estimates were least accurate for ears with severe hearing loss, and mild loss also showed larger errors than normal or moderate loss.
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
Many ears were excluded from threshold prediction because they failed signal-quality criteria, especially at 0.5 and 8 kHz, so the reported accuracy applies to the ears that passed. Under the linear method, 62.7%, 49.3%, 46.6%, 54.3%, and 62.0% of ears failed the inclusion criterion at 0.5, 1, 2, 4, and 8 kHz; under the neural-network method, 23.2%, 16.2%, 17.6%, 21.2%, and 34.7% failed. Prediction was least accurate for ears with severe hearing loss, and mild loss also showed larger errors than normal or moderate loss. The test took about 6 min per frequency, except 4 kHz where it took about 5.4 min, and the authors state that recording time must be shortened before clinical application. The study used standard octave frequencies only, and the authors note that notches in SFOAE spectra may reduce response level and signal-to-noise ratio, possibly leading to overestimated thresholds.
Declared interests
Funding was provided by the National Natural Science Foundation of China.
The easy way to misread this
Do not conclude that SFOAE testing can replace pure-tone audiometry. Many ears did not meet the signal-quality criteria, severe hearing loss was predicted least accurately, and the authors state that testing time must be shortened before clinical application.