Time-Frequency distributions of heart sound signals: A Comparative study using convolutional neural networks
Time-Frequency Distributions (TFDs) support the heart sound characterisation and classification in early cardiac screening. However, despite the frequent use of TFDs in signal analysis, no comprehensive study has been conducted to compare their performances in deep learning for automatic diagnosis....
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2023-06-01
|
Series: | Biomedical Engineering Advances |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2667099223000233 |
_version_ | 1797807211907383296 |
---|---|
author | Xinqi Bao Yujia Xu Hak-Keung Lam Mohamed Trabelsi Ines Chihi Lilia Sidhom Ernest N. Kamavuako |
author_facet | Xinqi Bao Yujia Xu Hak-Keung Lam Mohamed Trabelsi Ines Chihi Lilia Sidhom Ernest N. Kamavuako |
author_sort | Xinqi Bao |
collection | DOAJ |
description | Time-Frequency Distributions (TFDs) support the heart sound characterisation and classification in early cardiac screening. However, despite the frequent use of TFDs in signal analysis, no comprehensive study has been conducted to compare their performances in deep learning for automatic diagnosis. This study is the first to investigate and compare the optimal use of single/combined TFDs for heart sound classification using deep learning. The main contribution of this study is that it provides practical insights into the selection of TFDs as convolutional neural network (CNN) inputs and the design of CNN architecture for heart sound classification. The presented results revealed that: 1) The transformation of the heart sound signal into the TF domain achieves higher classification performance than using raw signal patterns as input. Overall, the difference in the performance was slight among the applied TFDs for all participated CNNs (within 1.3% in MAcc (average of sensitivity and specificity)). However, continuous wavelet transform (CWT) and Chirplet transform (CT) outperformed the rest (surpassing by approximately 0.5−1.3% in MAcc). 2) The appropriate increase of the CNN capacity and architecture optimisation can improve the performance, while the network architecture should not be overly complicated. Based on the results on ResNet or SEResNet, the increasing parameter number and the depth of the structure do not improve the performance apparently. 3) Combining TFDs as CNN inputs did not significantly improve the classification results. The results of this study provide valuable insights for researchers and practitioners in the field of automatic diagnosis of heart sounds with deep learning, particularly in selecting TFDs as CNN input and designing CNN architecture for heart sound classification. |
first_indexed | 2024-03-13T06:19:10Z |
format | Article |
id | doaj.art-efa68b70667640df9c3e75830068851b |
institution | Directory Open Access Journal |
issn | 2667-0992 |
language | English |
last_indexed | 2024-03-13T06:19:10Z |
publishDate | 2023-06-01 |
publisher | Elsevier |
record_format | Article |
series | Biomedical Engineering Advances |
spelling | doaj.art-efa68b70667640df9c3e75830068851b2023-06-10T04:28:54ZengElsevierBiomedical Engineering Advances2667-09922023-06-015100093Time-Frequency distributions of heart sound signals: A Comparative study using convolutional neural networksXinqi Bao0Yujia Xu1Hak-Keung Lam2Mohamed Trabelsi3Ines Chihi4Lilia Sidhom5Ernest N. Kamavuako6Corresponding author.; Department of Engineering, King’s College London, Strand, London, WC2R 2LS, United KingdomDepartment of Engineering, King’s College London, Strand, London, WC2R 2LS, United KingdomDepartment of Engineering, King’s College London, Strand, London, WC2R 2LS, United KingdomDepartment of Electronic and Communications Engineering, Kuwait College of Science and Technology, KuwaitDepartment of Engineering, Université du Luxembourg, 1359, LuxembourgNational Engineering School of Bizerta, Carthage University, Tunis, 2070, TunisiaDepartment of Engineering, King’s College London, Strand, London, WC2R 2LS, United Kingdom; Faculté de Médecine, Université de Kindu, Kindu, Democratic Republic of the CongoTime-Frequency Distributions (TFDs) support the heart sound characterisation and classification in early cardiac screening. However, despite the frequent use of TFDs in signal analysis, no comprehensive study has been conducted to compare their performances in deep learning for automatic diagnosis. This study is the first to investigate and compare the optimal use of single/combined TFDs for heart sound classification using deep learning. The main contribution of this study is that it provides practical insights into the selection of TFDs as convolutional neural network (CNN) inputs and the design of CNN architecture for heart sound classification. The presented results revealed that: 1) The transformation of the heart sound signal into the TF domain achieves higher classification performance than using raw signal patterns as input. Overall, the difference in the performance was slight among the applied TFDs for all participated CNNs (within 1.3% in MAcc (average of sensitivity and specificity)). However, continuous wavelet transform (CWT) and Chirplet transform (CT) outperformed the rest (surpassing by approximately 0.5−1.3% in MAcc). 2) The appropriate increase of the CNN capacity and architecture optimisation can improve the performance, while the network architecture should not be overly complicated. Based on the results on ResNet or SEResNet, the increasing parameter number and the depth of the structure do not improve the performance apparently. 3) Combining TFDs as CNN inputs did not significantly improve the classification results. The results of this study provide valuable insights for researchers and practitioners in the field of automatic diagnosis of heart sounds with deep learning, particularly in selecting TFDs as CNN input and designing CNN architecture for heart sound classification.http://www.sciencedirect.com/science/article/pii/S2667099223000233Heart soundTime-frequency distributionsConvolutional neural networks |
spellingShingle | Xinqi Bao Yujia Xu Hak-Keung Lam Mohamed Trabelsi Ines Chihi Lilia Sidhom Ernest N. Kamavuako Time-Frequency distributions of heart sound signals: A Comparative study using convolutional neural networks Biomedical Engineering Advances Heart sound Time-frequency distributions Convolutional neural networks |
title | Time-Frequency distributions of heart sound signals: A Comparative study using convolutional neural networks |
title_full | Time-Frequency distributions of heart sound signals: A Comparative study using convolutional neural networks |
title_fullStr | Time-Frequency distributions of heart sound signals: A Comparative study using convolutional neural networks |
title_full_unstemmed | Time-Frequency distributions of heart sound signals: A Comparative study using convolutional neural networks |
title_short | Time-Frequency distributions of heart sound signals: A Comparative study using convolutional neural networks |
title_sort | time frequency distributions of heart sound signals a comparative study using convolutional neural networks |
topic | Heart sound Time-frequency distributions Convolutional neural networks |
url | http://www.sciencedirect.com/science/article/pii/S2667099223000233 |
work_keys_str_mv | AT xinqibao timefrequencydistributionsofheartsoundsignalsacomparativestudyusingconvolutionalneuralnetworks AT yujiaxu timefrequencydistributionsofheartsoundsignalsacomparativestudyusingconvolutionalneuralnetworks AT hakkeunglam timefrequencydistributionsofheartsoundsignalsacomparativestudyusingconvolutionalneuralnetworks AT mohamedtrabelsi timefrequencydistributionsofheartsoundsignalsacomparativestudyusingconvolutionalneuralnetworks AT ineschihi timefrequencydistributionsofheartsoundsignalsacomparativestudyusingconvolutionalneuralnetworks AT liliasidhom timefrequencydistributionsofheartsoundsignalsacomparativestudyusingconvolutionalneuralnetworks AT ernestnkamavuako timefrequencydistributionsofheartsoundsignalsacomparativestudyusingconvolutionalneuralnetworks |