Subscribe to RSS
![](/products/assets/desktop/img/oa-logo.png)
DOI: 10.1055/s-0040-1711222
Using intelligent algorithms on singlesweeps to evaluate thresholds and waveforms of brainstem evoked response audiometry (BERA) on animal model
Introduction Unrecognized changes in hearing perception and their processing pose a serious barrier to integration into our society. The threshold of hearing level from the averaged measurements of brain stem potentials is visually determined in clinical practice. If in addition to the hearing threshold, questions arise in regard of wave morphology , it is not possible to quantify these features objectively without having, concrete references and evaluation algorithms.
Material / Methods As part of hearing measurements on mice with genetically modified calcium channels on the hair cells, single sweep ABR measurements on 30 mice of a control group of normal hearing animals were performed. Based on a top-down process, the algorithm was trained on the single sweeps to distinguish evoked responses from neuronal noise. Furthermore, automatically acquired features from the singles weeps were compared with each other via a cluster analysis.
Results The trained algorithm detects the hearing threshold very reliably with a maximum deviation of +/- 5dB from the visual estimate. In addition, significant differences in the wave growth function of "normal hearing" can be demonstrated. By means of the cluster analysis, opposing effects in the single sweeps between immediately consecutive wave forms of a sweep can be visualized. If a wave peak in the single sweep is pronounced, the subsequent wave peak is less pronounced in the same single sweep and vice versa.
Poster-PDF A-1157.PDF
Publication History
Article published online:
10 June 2020
© 2020. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial-License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (https://creativecommons.org/licenses/by-nc-nd/4.0/).
© Georg Thieme Verlag KG
Stuttgart · New York