Automatic Diagnosis of Voiding Dysfunction From Sound Signal

The aim of this paper is to present the results of an experiment towards Sonouroflowmetry, a novel approach for recognition of potential voiding dysfunctions based on machine learning classification of sound records that are obtained while a patient urinates into water in a toilet bowl. Such approach could enable a diagnosis of the voiding dysfunctions via a mobile device. We provide a comparison of 69 state-of-the-art classification methods.