Method and apparatus for training and evaluating artificial neural networks used to determine lung pathology

Abstract

A computer-implemented method for determining lung pathology from an audio respiratory signal comprises inputting a plurality of audio files comprising a training set into an artificial neural network (ANN), wherein the plurality of audio files comprise sessions with patients with known pathologies of known degrees of severity. The method further comprises annotating the plurality of audio files with metadata relevant to the patients and the known pathologies and analyzing the plurality of audio files, wherein the analyzing comprises extracting spectrograms for each of the plurality of audio files and a plurality of descriptors associated with wheeze and crackle from the plurality of audio files. Additionally, the method comprises training the ANN using the plurality of audio files, the spectrograms, the metadata and the plurality of descriptors. The method finally comprises determining a lung pathology associated with a new sound recording inputted into the ANN.