Wheeze Events Detection Using Convolutional Recurrent Neural Network

Hakki L., SERBES G.

2023 Innovations in Intelligent Systems and Applications Conference, ASYU 2023, Sivas, Turkey, 11 - 13 October 2023 identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/asyu58738.2023.10296661
  • City: Sivas
  • Country: Turkey
  • Keywords: Convolutional Neural Networks, Deep Learning, Recurrent Neural Networks, Respiratory Sounds, Short Time Fourier Transform, Wheezes
  • Yıldız Technical University Affiliated: Yes


Chronic respiratory disorders (CRDs) affect the airways and other structures in the lungs. According to WHO, CRDs are a major cause of death globally. Early diagnosis and monitoring of individuals with respiratory disorders are crucial due to the severity and prevalence of these disorders. Auscultation is a common method used to diagnose respiratory patients. However, the classical auscultation procedure has some limitations, such as being subjective, depending on the physician's expertise, and being inaccurate in noisy environments. To tackle those limitations, this project aims to implement a method for the detection of adventitious respiratory sounds, particularly wheeze sounds, using data derived from ICBHI open data. Short-time Fourier transforms (STFT) of the audio data were applied for the feature extraction. The system was implemented to perform wheeze sound detection using a recurrent neural network (RNN) based deep-learning model.