Automated Lung Sound Classification Using a Hybrid CNN-LSTM Network and Focal Loss Function
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , , , , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | http://hdl.handle.net/10316/103499 https://doi.org/10.3390/s22031232 |
Resumo: | Respiratory diseases constitute one of the leading causes of death worldwide and directly affect the patient's quality of life. Early diagnosis and patient monitoring, which conventionally include lung auscultation, are essential for the efficient management of respiratory diseases. Manual lung sound interpretation is a subjective and time-consuming process that requires high medical expertise. The capabilities that deep learning offers could be exploited in order that robust lung sound classification models can be designed. In this paper, we propose a novel hybrid neural model that implements the focal loss (FL) function to deal with training data imbalance. Features initially extracted from short-time Fourier transform (STFT) spectrograms via a convolutional neural network (CNN) are given as input to a long short-term memory (LSTM) network that memorizes the temporal dependencies between data and classifies four types of lung sounds, including normal, crackles, wheezes, and both crackles and wheezes. The model was trained and tested on the ICBHI 2017 Respiratory Sound Database and achieved state-of-the-art results using three different data splitting strategies-namely, sensitivity 47.37%, specificity 82.46%, score 64.92% and accuracy 73.69% for the official 60/40 split, sensitivity 52.78%, specificity 84.26%, score 68.52% and accuracy 76.39% using interpatient 10-fold cross validation, and sensitivity 60.29% and accuracy 74.57% using leave-one-out cross validation. |
id |
RCAP_56ceb904d029f2a4d42b0e9291ae189a |
---|---|
oai_identifier_str |
oai:estudogeral.uc.pt:10316/103499 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Automated Lung Sound Classification Using a Hybrid CNN-LSTM Network and Focal Loss Functionlung soundscrackleswheezesSTFTCNNLSTMfocal lossCOPDasthmaAuscultationHumansLungNeural Networks, ComputerQuality of LifeRespiratory SoundsRespiratory diseases constitute one of the leading causes of death worldwide and directly affect the patient's quality of life. Early diagnosis and patient monitoring, which conventionally include lung auscultation, are essential for the efficient management of respiratory diseases. Manual lung sound interpretation is a subjective and time-consuming process that requires high medical expertise. The capabilities that deep learning offers could be exploited in order that robust lung sound classification models can be designed. In this paper, we propose a novel hybrid neural model that implements the focal loss (FL) function to deal with training data imbalance. Features initially extracted from short-time Fourier transform (STFT) spectrograms via a convolutional neural network (CNN) are given as input to a long short-term memory (LSTM) network that memorizes the temporal dependencies between data and classifies four types of lung sounds, including normal, crackles, wheezes, and both crackles and wheezes. The model was trained and tested on the ICBHI 2017 Respiratory Sound Database and achieved state-of-the-art results using three different data splitting strategies-namely, sensitivity 47.37%, specificity 82.46%, score 64.92% and accuracy 73.69% for the official 60/40 split, sensitivity 52.78%, specificity 84.26%, score 68.52% and accuracy 76.39% using interpatient 10-fold cross validation, and sensitivity 60.29% and accuracy 74.57% using leave-one-out cross validation.MDPI2022-02-06info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlehttp://hdl.handle.net/10316/103499http://hdl.handle.net/10316/103499https://doi.org/10.3390/s22031232eng1424-8220Petmezas, GeorgiosCheimariotis, Grigorios-ArisStefanopoulos, LeandrosRocha, BrunoPaiva, Rui PedroKatsaggelos, Aggelos K.Maglaveras, Nicosinfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2022-11-16T21:35:57Zoai:estudogeral.uc.pt:10316/103499Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T21:20:19.490857Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Automated Lung Sound Classification Using a Hybrid CNN-LSTM Network and Focal Loss Function |
title |
Automated Lung Sound Classification Using a Hybrid CNN-LSTM Network and Focal Loss Function |
spellingShingle |
Automated Lung Sound Classification Using a Hybrid CNN-LSTM Network and Focal Loss Function Petmezas, Georgios lung sounds crackles wheezes STFT CNN LSTM focal loss COPD asthma Auscultation Humans Lung Neural Networks, Computer Quality of Life Respiratory Sounds |
title_short |
Automated Lung Sound Classification Using a Hybrid CNN-LSTM Network and Focal Loss Function |
title_full |
Automated Lung Sound Classification Using a Hybrid CNN-LSTM Network and Focal Loss Function |
title_fullStr |
Automated Lung Sound Classification Using a Hybrid CNN-LSTM Network and Focal Loss Function |
title_full_unstemmed |
Automated Lung Sound Classification Using a Hybrid CNN-LSTM Network and Focal Loss Function |
title_sort |
Automated Lung Sound Classification Using a Hybrid CNN-LSTM Network and Focal Loss Function |
author |
Petmezas, Georgios |
author_facet |
Petmezas, Georgios Cheimariotis, Grigorios-Aris Stefanopoulos, Leandros Rocha, Bruno Paiva, Rui Pedro Katsaggelos, Aggelos K. Maglaveras, Nicos |
author_role |
author |
author2 |
Cheimariotis, Grigorios-Aris Stefanopoulos, Leandros Rocha, Bruno Paiva, Rui Pedro Katsaggelos, Aggelos K. Maglaveras, Nicos |
author2_role |
author author author author author author |
dc.contributor.author.fl_str_mv |
Petmezas, Georgios Cheimariotis, Grigorios-Aris Stefanopoulos, Leandros Rocha, Bruno Paiva, Rui Pedro Katsaggelos, Aggelos K. Maglaveras, Nicos |
dc.subject.por.fl_str_mv |
lung sounds crackles wheezes STFT CNN LSTM focal loss COPD asthma Auscultation Humans Lung Neural Networks, Computer Quality of Life Respiratory Sounds |
topic |
lung sounds crackles wheezes STFT CNN LSTM focal loss COPD asthma Auscultation Humans Lung Neural Networks, Computer Quality of Life Respiratory Sounds |
description |
Respiratory diseases constitute one of the leading causes of death worldwide and directly affect the patient's quality of life. Early diagnosis and patient monitoring, which conventionally include lung auscultation, are essential for the efficient management of respiratory diseases. Manual lung sound interpretation is a subjective and time-consuming process that requires high medical expertise. The capabilities that deep learning offers could be exploited in order that robust lung sound classification models can be designed. In this paper, we propose a novel hybrid neural model that implements the focal loss (FL) function to deal with training data imbalance. Features initially extracted from short-time Fourier transform (STFT) spectrograms via a convolutional neural network (CNN) are given as input to a long short-term memory (LSTM) network that memorizes the temporal dependencies between data and classifies four types of lung sounds, including normal, crackles, wheezes, and both crackles and wheezes. The model was trained and tested on the ICBHI 2017 Respiratory Sound Database and achieved state-of-the-art results using three different data splitting strategies-namely, sensitivity 47.37%, specificity 82.46%, score 64.92% and accuracy 73.69% for the official 60/40 split, sensitivity 52.78%, specificity 84.26%, score 68.52% and accuracy 76.39% using interpatient 10-fold cross validation, and sensitivity 60.29% and accuracy 74.57% using leave-one-out cross validation. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-02-06 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/10316/103499 http://hdl.handle.net/10316/103499 https://doi.org/10.3390/s22031232 |
url |
http://hdl.handle.net/10316/103499 https://doi.org/10.3390/s22031232 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
1424-8220 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.publisher.none.fl_str_mv |
MDPI |
publisher.none.fl_str_mv |
MDPI |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799134095917187072 |