Objective hearing threshold identification from auditory brainstem response measurements using supervised and self-supervised approaches

Hearing loss is a major health problem and psychological burden in humans. Mouse models offer a possibility to elucidate genes involved in the underlying developmental and pathophysiological mechanisms of hearing impairment. To this end, large-scale mouse phenotyping programs include auditory phenotyping of single-gene knockout mouse lines. Using the auditory brainstem response (ABR) procedure, the German Mouse Clinic and similar facilities worldwide have produced large, uniform data sets of averaged ABR raw data of mutant and wildtype mice. In the course of standard ABR analysis, hearing thresholds are assessed visually by trained staff from series of signal curves of increasing sound pressure level. This is time-consuming and prone to be biased by the reader as well as the graphical display quality and scale. In an attempt to reduce workload and improve quality and reproducibility, we developed and compared two methods for automated hearing threshold identification from averaged ABR raw data: a supervised approach involving two combined neural networks trained on human-generated labels and a self-supervised approach, which exploits the signal power spectrum and combines random forest sound level estimation with a piece-wise curve fitting algorithm for threshold finding. We show that both models work well, outperform human threshold detection, and are suitable for fast, reliable, and unbiased hearing threshold detection and quality control. In a high-throughput mouse phenotyping environment, both methods perform well as part of an automated end-to-end screening pipeline to detect candidate genes for hearing involvement. Code for both models as well as data used for this work are freely available.

[1]  Gaye Lightbody,et al.  Auditory brainstem response classification: A hybrid model using time and frequency features , 2007, Artif. Intell. Medicine.

[2]  Bo Li,et al.  Automated Threshold Determination of Auditory Evoked Brainstem Responses by Cross-correlation Analysis with Varying Sweep Number , 2019 .

[3]  Harvey Dillon,et al.  The detection of infant cortical auditory evoked potentials (CAEPs) using statistical and visual detection techniques. , 2010, Journal of the American Academy of Audiology.

[4]  G. Camp,et al.  The hereditary hearing loss homepage , 1997 .

[5]  David Parker,et al.  Auditory brainstem response threshold estimation: subjective threshold estimation by experienced clinicians in a computer simulation of the clinical test , 2004, International journal of audiology.

[6]  K D Wernecke,et al.  Objective detection of auditory brainstem potentials: Comparison of statistical tests in the time and frequency domains , 2000, Scandinavian audiology.

[7]  D. Simpson,et al.  Objective measures for detecting the auditory brainstem response: comparisons of specificity, sensitivity and detection time , 2018, International journal of audiology.

[8]  S. Arnold Objective versus Visual Detection of the Auditory Brain Stem Response , 1985, Ear and hearing.

[9]  B. A.,et al.  Disease model discovery from 3,328 gene knockouts by The International Mouse Phenotyping Consortium , 2018, Yearbook of Paediatric Endocrinology.

[10]  Richard M McKearney,et al.  Objective auditory brainstem response classification using machine learning , 2019, International journal of audiology.

[11]  Steve D. M. Brown,et al.  High-throughput discovery of novel developmental phenotypes , 2017 .

[12]  K. Steel,et al.  Using the Auditory Brainstem Response (ABR) to Determine Sensitivity of Hearing in Mutant Mice , 2011, Current protocols in mouse biology.

[13]  Jing Lv,et al.  Objective detection of evoked potentials using a bootstrap technique. , 2007, Medical engineering & physics.

[14]  Judith A. Blake,et al.  Mouse Genome Database (MGD) 2019 , 2018, Nucleic Acids Res..

[15]  Steve D. M. Brown,et al.  A large scale hearing loss screen reveals an extensive unexplored genetic landscape for auditory dysfunction , 2017, Nature Communications.

[16]  Qiuju Wang,et al.  Automatic Recognition of Auditory Brainstem Response Characteristic Waveform Based on Bidirectional Long Short-Term Memory , 2021, Frontiers in Medicine.

[17]  J. Clements,et al.  Automated threshold detection for auditory brainstem responses: comparison with visual estimation in a stem cell transplantation study , 2009, BMC Neuroscience.

[18]  Achim Schilling,et al.  Objective Estimation of Sensory Thresholds Based on Neurophysiological Parameters , 2018, Front. Neurosci..

[19]  Abdul Hamid Bin Adom,et al.  A machine learning approach for distinguishing hearing perception level using auditory evoked potentials , 2014, 2014 IEEE Conference on Biomedical Engineering and Sciences (IECBES).

[20]  K. G. Hill,et al.  Auditory brainstem response in tammar wallaby (Macropus eugenii) , 1997, Hearing Research.

[21]  Ewelina Majda-Zdancewicz,et al.  Classification of auditory brainstem response using wavelet decomposition and SVM network , 2016 .

[22]  R A Dobie,et al.  Analysis of auditory evoked potentials by magnitude-squared coherence. , 1989, Ear and hearing.

[24]  Nurettin Acir,et al.  Automatic classification of auditory brainstem responses using SVM-based feature selection algorithm for threshold detection , 2006, Eng. Appl. Artif. Intell..

[25]  O. Ozdamar,et al.  Determining hearing threshold from brain stem evoked potentials. Optimizing a neural network to improve classification performance , 1994, IEEE Engineering in Medicine and Biology Magazine.

[26]  O Ozdamar,et al.  Computer methods for on-line hearing testing with auditory brain stem responses. , 1990, Ear and hearing.

[27]  D. Kolbe,et al.  Insights into the Biology of Hearing and Deafness Revealed by Single-Cell RNA Sequencing , 2019, Cell reports.

[28]  Jean-François Motsch,et al.  Objective detection of brainstem auditory evoked potentials with a priori information from higher presentation levels , 2002, Artif. Intell. Medicine.

[29]  Huiru Zheng,et al.  A Comparison of Supervised Classification Methods for Auditory Brainstem Response Determination , 2007, MedInfo.

[30]  François Schiettecatte,et al.  OMIM.org: Online Mendelian Inheritance in Man (OMIM®), an online catalog of human genes and genetic disorders , 2014, Nucleic Acids Res..

[31]  O Ozdamar,et al.  Automated electrophysiologic hearing testing using a threshold-seeking algorithm. , 1994, Journal of the American Academy of Audiology.

[32]  Jacqueline K. White,et al.  Mouse screen reveals multiple new genes underlying mouse and human hearing loss , 2019, PLoS biology.

[33]  N. Cliff Dominance statistics: Ordinal analyses to answer ordinal questions. , 1993 .

[34]  M. Charles Liberman,et al.  A simple algorithm for objective threshold determination of auditory brainstem responses , 2019, Hearing Research.

[35]  D. Corey,et al.  Gene Expression by Mouse Inner Ear Hair Cells during Development , 2015, The Journal of Neuroscience.

[36]  Arne Leijon,et al.  Analysis of Click-Evoked Auditory Brainstem Responses Using Time Domain Cross-Correlations Between Interleaved Responses , 2014, Ear and hearing.

[37]  Werner Müller,et al.  Introducing the German Mouse Clinic: open access platform for standardized phenotyping , 2005, Nature Methods.

[38]  H. Fuchs,et al.  The German Mouse Clinic: a platform for systemic phenotype analysis of mouse models. , 2009, Current pharmaceutical biotechnology.

[39]  Sang W Shin,et al.  Hearing Loss in Adults. , 2018, The New England journal of medicine.