Abstract

The recognition of various lung sounds recorded using electronic stethoscopes plays a significant role in the early diagnoses of respiratory diseases. To increase the accuracy of specialist evaluations, machine learning techniques have been intensely employed during the past 30 years. In the current study, a new pretrained Convolutional Neural Network (CNN) model is proposed for the extraction of deep features. In the CNN architecture, an average-pooling layer and a max-pooling layer are connected in parallel in order to boost classification performance. The deep features are utilized as the input of the Linear Discriminant Analysis (LDA) classifier using the Random Subspace Ensembles (RSE) method. The proposed method was evaluated against a challenge dataset known as ICBHI 2017. The deep features and the LDA with RSE method provided the best accuracy score when compared to other existing methods using the same dataset, improving the classification accuracy by 5.75%.

Highlights

  • Lung disease ranks third among fatality causes worldwide

  • In [8], the features extracted from time-frequency and timescale analysis methods are utilized for the detection of normal lung sounds and crackles, with k-Nearest Neighbors (k-NN), Multilayer Perceptron (MLP) and Support Vector Machine (SVM) used for the classification stage

  • In [14], a Convolutional Neural Network (CNN) model was employed for the classification of lung sounds, with the CNN shown to perform superior to MelFrequency Cepstral Coefficients (MFCCs) features in the SVM

Read more

Summary

INTRODUCTION

Lung disease ranks third among fatality causes worldwide. According to the World Health Organization (WHO), more than 3 million people die each year due to respiratory diseases [1]. Auscultation is a method by which doctors evaluate and diagnose lung diseases using a stethoscope It is known as a low-cost, easy to apply, and reliable test that requires minimal diagnosis duration [5]. In [8], the features extracted from time-frequency and timescale analysis methods are utilized for the detection of normal lung sounds and crackles, with k-Nearest Neighbors (k-NN), Multilayer Perceptron (MLP) and SVM used for the classification stage. In [14], a CNN model was employed for the classification of lung sounds, with the CNN shown to perform superior to MFCC features in the SVM. In the method proposed in the current study, a hybrid approach was applied in order to increase the classification performance in the identification of lung sounds. In the Conclusions section, the experimental results are interpreted according to performance criteria and other methods that have used the same dataset

THE METHODOLOGY
SPECTROGRAM IMAGES
LINEAR DISCRİMİNANT ANALYSİS
RANDOM SUBSPACE ENSEMBLES
EXPERİMENTAL SETUP AND RESULTS
Findings
CONCLUSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call