Convolutional Neural Network-based Model for Lung Sounds Classification
Authors:
- 9 August 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 555-558
Abstract
Deep learning approaches are gaining popularity in the medical field for diagnostics and predictive analytics. This study aims to improve the classification of respiratory sounds recorded by electronic stethoscopes using a deep approach based on a convolutional neural network (CNN). The proposed scheme’s performance is studied using the largest publicly available lung sound dataset (ICBHI 2017). We implement variant frequency representations, including CQT, STFT, and Mel-STFT, combined with Empirical mode decomposition (EMD). We propose the use of data normalization on top of data augmentation techniques. Furthermore, to increases the model performance, we used hyperparameters optimization and regulators. Experimental results show that our method outperforms competing methods on anomaly prediction tasks.Keywords
This publication has 13 references indexed in Scilit:
- Classification of Lung Sounds With CNN Model Using Parallel Pooling StructureIEEE Access, 2020
- Detecting Respiratory Pathologies Using Convolutional Neural Networks and Variational Autoencoders for Unbalancing DataSensors, 2020
- Convolutional neural networks based efficient approach for classification of lung diseasesHealth Information Science and Systems, 2019
- Detection of Adventitious Respiratory Sounds based on Convolutional Neural Network2019
- Deep Auscultation: Predicting Respiratory Anomalies and Diseases via Recurrent Neural Networks2019
- An open access database for the evaluation of respiratory sound classification algorithmsPhysiological Measurement, 2019
- Convolutional Neural Networks Learning from Respiratory data2018
- Classification of lung sounds using convolutional neural networksEURASIP Journal on Image and Video Processing, 2017
- Estimation of instantaneous frequency from empirical mode decomposition on respiratory sounds analysis2013
- Respiratory sounds classification using cepstral analysis and Gaussian mixture models2007