Detection of transient evoked otoacoustic emissions by adaptive signal enhancement

Otoacoustic emissions (OAEs) has been considered as an excellent objective tool in clinics for diagnosing hearing loss. The signal-to-noise ratio (SNR) and correlation coefficient of OAEs are very important for the purpose of diagnosis. An adaptive signal enhancer (ASE) based on the Least Mean Square (LMS) algorithm is presented to detect transient evoked OAEs (TEOAEs). The TEOAEs detection results from 106 ears show that ASE reaches better estimation of TEOAEs than a conventional ensemble averaging (EA) technique. With the ASE, the improvement of SNR was increased faster than that with the EA and the number of sweeps required can be markedly reduced. The detection time with ASE could be shortened by about 50% in comparison with that of EA.

Correlation among thresholds of evoked otoacoustic emissions,behavioral responses and auditory brainstem responses

In 35 adult human subjects (58 ears) thresholds of rarefaction click evoked otoacoustic emission (EOAE), behavioral response (BR) and auditory brainstem response (ABR) were measured and compared, and correlation coefficients (r) among them calculated. The results revealed that 86 % of the measured thresholds were in the range from 10 to 45 dB (nHL) for EOAE, 88 % from 10 to 25 dB for BR and 88 % from 10 to 35 for ABR. The correlation coefficients for the whole threshold sample were 0.415 (p< 0.002) for EOAE vs. BR, 0.501 (p< 0.001) for EOAE vs. ABR and 0.702 (p< 0.001) for ABR vs. BR, all indicating highly significant correlation. However, for those ears whose BR, ABR or EOAE thresholds were elevated, equaling to or exceeding 25 dB, there is no significant correlation between thresholds of EOAE and BR and of EOAE and ABR (r range: 0.176-0.310, p>0.05). In contrast, significant correlation between BR and ABR thresholds in the same conditions still remained (r: 0.533-0.720, p<0.05). The experimental results indicate that the correlation between EOAE and hearing thresholds is significant only for subjects with normal or slightly lost hearing. EOAE is valuable in audiometry only as an accessory tool.

Otoacoustic emissions value in patients with idiopathic sudden sensorineural hearing loss

Objectives:This study aimed to determine the prognostic value of otoacoustic emissions(OAEs)in idiopathic sudden sensorineural hearing loss patients.Methods:The study included 30 subjects with unilateral idiopathic sudden sensorineural hearing loss(ISSNHL).Each patient was evaluated four times:at baseline and after one week,one month,and three months of treatment.During each visit,each patient was subjected to full audiological history,otoscopic examination,basic audiological evaluations,and transiently evoked and distortion product otoacoustic emission(TEOAEs&DEOAEs).Results:The hearing thresholds(frequency range 250e8000 Hz)and word recognition scores of patients with detectable TEOAEs and DPOAEs improved significantly,whereas no significant improvements were observed in those with no response.Conclusion:Hearing improvement is better in patients with detectable TEOAEs and DPOAEs.As a result,TEOAEs and DPOAEs are recommended as routine tests in all SSNHL patients to predict outcomes and monitor treatment as TEOAEs and DPOAEs reflect the cochlear OHCs activity.

Efficient Authentication System Using Wavelet Embeddings of Otoacoustic Emission Signals

Biometrics,which has become integrated with our daily lives,could fall prey to falsification attacks,leading to security concerns.In our paper,we use Transient Evoked Otoacoustic Emissions(TEOAE)that are generated by the human cochlea in response to an external sound stimulus,as a biometric modality.TEOAE are robust to falsification attacks,as the uniqueness of an individual’s inner ear cannot be impersonated.In this study,we use both the raw 1D TEOAE signals,as well as the 2D time-frequency representation of the signal using Continuous Wavelet Transform(CWT).We use 1D and 2D Convolutional Neural Networks(CNN)for the former and latter,respectively,to derive the feature maps.The corresponding lower-dimensional feature maps are obtained using principal component analysis,which is then used as features to build classifiers using machine learning techniques for the task of person identification.T-SNE plots of these feature maps show that they discriminate well among the subjects.Among the various architectures explored,we achieve a best-performing accuracy of 98.95%and 100%using the feature maps of the 1D-CNN and 2D-CNN,respectively,with the latter performance being an improvement over all the earlier works.This performance makes the TEOAE based person identification systems deployable in real-world situations,along with the added advantage of robustness to falsification attacks.