Automatic auscultation classification of abnormal lung sounds in critical patients through deep learning models

Yu Sheng Wu, Chia Hung Liao, Shyan Ming Yuan*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This research aims to use the output signals of a stethoscope and classify them through deep learning models automatically. In this research, the dataset consists of four classes, normal, wheezing, crackles, and unknown are used. To effectively classify each signal, we use the spectrogram generated by the short-time fast Fourier transform as the feature value of each lung sound signal and found the best parameters to do model selection. Besides, we also adopt Depthwise separable (DS) convolution technic, and refer to the architecture of Mobile-Net, to achieve the purpose of high accuracy and low model parameters.

Original languageEnglish
Title of host publicationProceedings of the 3rd IEEE International Conference on Knowledge Innovation and Invention 2020, ICKII 2020
EditorsTeen-Hang Meen
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages9-11
Number of pages3
ISBN (Electronic)9781728193335
DOIs
StatePublished - 21 Aug 2020
Event3rd IEEE International Conference on Knowledge Innovation and Invention, ICKII 2020 - Kaohsiung, Taiwan
Duration: 21 Aug 202023 Aug 2020

Publication series

NameProceedings of the 3rd IEEE International Conference on Knowledge Innovation and Invention 2020, ICKII 2020

Conference

Conference3rd IEEE International Conference on Knowledge Innovation and Invention, ICKII 2020
CountryTaiwan
CityKaohsiung
Period21/08/2023/08/20

Keywords

  • Adventitious lung sound
  • Auscultations
  • Convolutional neural network
  • Deep learning
  • Feature extraction

Fingerprint Dive into the research topics of 'Automatic auscultation classification of abnormal lung sounds in critical patients through deep learning models'. Together they form a unique fingerprint.

Cite this