Computational analysis based on audioprofiles: A new possibility for patient stratification in office-based otology

  • Oren Weininger Department of Otolaryngology, Hannover Medical School, Hannover, Germany.
  • Athanasia Warnecke | warnecke.athanasia@mh-hannover.de Department of Otolaryngology, Hannover Medical School, Hannover; Cluster of Excellence, Hearing4all German Research Foundation, Hannover, Germany.
  • Anke Lesinski-Schiedat Department of Otolaryngology, Hannover Medical School, Hannover, Germany.
  • Thomas Lenarz Department of Otolaryngology, Hannover Medical School, Hannover; Cluster of Excellence, Hearing4all German Research Foundation, Hannover, Germany.
  • Stefan Stolle Department of Otolaryngology, Hannover Medical School, Hannover, Germany.

Abstract

Genetic contribution to progressive hearing loss in adults is underestimated. Established machine learning-based software could offer a rapid supportive tool to stratify patients with progressive hearing loss. A retrospective longitudinal analysis of 141 adult patients presenting with hearing loss was performed. Hearing threshold was measured at least twice 18 months or more apart. Based on the baseline audiogram, hearing thresholds and age were uploaded to AudioGene v4® (Center for Bioinformatics and Computational Biology at The University of Iowa City, IA, USA) to predict the underlying genetic cause of hearing loss and the likely progression of hearing loss. The progression of hearing loss was validated by comparison with the most recent audiogram data of the patients. The most frequently predicted loci were DFNA2B, DFNA9 and DFNA2A. The frequency of loci/genes predicted by AudioGene remains consistent when using the initial or the final audiogram of the patients. In conclusion, machine learning-based software analysis of clinical data might be a useful tool to identify patients at risk for having autosomal dominant hearing loss. With this approach, patients with suspected progressive hearing loss could be subjected to close audiological followup, genetic testing and improved patient counselling.

Downloads

Download data is not yet available.
Published
2019-11-05
Section
Original Articles
Keywords:
Machine learning, Progressive hearing loss, Audiogram, Phenotype, Genotype
Statistics
Abstract views: 37

PDF: 30
Share it

PlumX Metrics

PlumX Metrics provide insights into the ways people interact with individual pieces of research output (articles, conference proceedings, book chapters, and many more) in the online environment. Examples include, when research is mentioned in the news or is tweeted about. Collectively known as PlumX Metrics, these metrics are divided into five categories to help make sense of the huge amounts of data involved and to enable analysis by comparing like with like.

How to Cite
Weininger, O., Warnecke, A., Lesinski-Schiedat, A., Lenarz, T., & Stolle, S. (2019). Computational analysis based on audioprofiles: A new possibility for patient stratification in office-based otology. Audiology Research, 9(2). https://doi.org/10.4081/audiores.2019.230