S. Hatzopoulos et al., CLINICAL APPLICABILITY OF TRANSIENT EVOKED OTOACOUSTIC EMISSIONS - IDENTIFICATION AND CLASSIFICATION OF HEARING-LOSS, Audiology & neuro-otology, 3(6), 1998, pp. 402-418
The study aimed at the development of a clinically applicable methodol
ogy that could: (1) discriminate transient evoked otoacoustic emission
(TEOAE) recordings from normal hearing or hearing impaired individual
s; (2) classify the nature of the hearing loss as conductive or as coc
hlear, and (3) define clear-cut TEOAE clinical criteria, A classificat
ion algorithm based on a multivariate discriminant analysis of fast Fo
urier transform data from recordings evoked by click stimuli of 50 +/-
2, 62 +/- 2, 68 +/- 2 and 80 +/- 2 dB SPL was used to discriminate 30
2 normal subjects from 383 subjects suffering from mild to moderate he
aring losses, The best discriminant model (QDF80) produced a sensitivi
ty of 93.8% and a specificity of 79.4%. When extra correlation criteri
a were serially applied to the classification outcome, the specificity
was increased to 85.3%, but the sensitivity was marginally decreased
to 91.7%. The classification of the correctly identified hearing-impai
red cases yielded 93.8% identification of conductive and 75.1% identif
ication of cochlear cases, A sensitivity analysis of the misclassified
hearing-impaired cases suggested that the TEOAE spectra are well corr
elated with the 2-kHz but poorly correlated with the 4-kHz octave freq
uency.