Diagnosing Noise-Induced Hearing Loss Sustained During Military Service Using Deep Neural Networks

Authors: Brian CJ Moore1*; Josef Schlittenlacher2

1University of Cambridge

2University College London


Background The diagnosis of noise-induced hearing loss (NIHL) is based on three requirements: a history of exposure to noise with the potential to cause hearing loss; the absence of known causes of hearing loss other than noise exposure; and the presence of certain features in the audiogram. All current methods for diagnosing NIHL have involved examination of the typical features of the audiograms of noise-exposed individuals and the formulation of rules for the identification of those features. This paper describes an alternative approach based on the use of multi-layer perceptrons (MLPs).

Method The development and evaluation of the MLPs was based on databases containing the ages and audiograms of individuals claiming compensation for NIHL sustained during military service (M-NIHL), who were assumed mostly to have M-NIHL, and control databases based on individuals with no known exposure to intense sounds. Rather than relying on intuitions to identify the features of the audiogram that characterize M-NIHL, the MLPs were trained so as to classify individuals as belonging to the exposed or control group based on their audiograms and ages, thereby automatically identifying the features of the audiogram that provide optimal classification. Two databases (noise-exposed and non-exposed) were used for training and validation of the MLPs and two independent databases were used for evaluation and further analyses.

Results The best-performing MLP was one trained to identify whether or not an individual had M-NIHL based on age and the audiogram for both ears. For the databases independent of those used for training of the MLP, this achieved a sensitivity of 0.986 and a specificity of 0.902, giving an overall accuracy markedly higher than for previous methods.

Conclusion The trained MLP achieved greater accuracy that previous methods in diagnosing whether an individual had M-NIHL.