Impact of Lung Segmentation on the Diagnosis and Explanation of COVID-19 in Chest X-ray Images

COVID-19 frequently provokes pneumonia, which can be diagnosed using imaging exams. Chest X-ray (CXR) is often useful because it is cheap, fast, widespread, and uses less radiation. Here, we demonstrate the impact of lung segmentation in COVID-19 identification using CXR images and evaluate which co...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Lucas O. Teixeira, Rodolfo M. Pereira, Diego Bertolini, Luiz S. Oliveira, Loris Nanni, George D. C. Cavalcanti, Yandre M. G. Costa
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	COVID-19 chest X-ray semantic segmentation explainable artificial intelligence Chemical technology TP1-1185
Acceso en línea:	https://doaj.org/article/a75d2b90afe9475282305ab023dd4324
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:a75d2b90afe9475282305ab023dd4324
record_format	dspace
spelling	oai:doaj.org-article:a75d2b90afe9475282305ab023dd43242021-11-11T19:07:23ZImpact of Lung Segmentation on the Diagnosis and Explanation of COVID-19 in Chest X-ray Images10.3390/s212171161424-8220https://doaj.org/article/a75d2b90afe9475282305ab023dd43242021-10-01T00:00:00Zhttps://www.mdpi.com/1424-8220/21/21/7116https://doaj.org/toc/1424-8220COVID-19 frequently provokes pneumonia, which can be diagnosed using imaging exams. Chest X-ray (CXR) is often useful because it is cheap, fast, widespread, and uses less radiation. Here, we demonstrate the impact of lung segmentation in COVID-19 identification using CXR images and evaluate which contents of the image influenced the most. Semantic segmentation was performed using a U-Net CNN architecture, and the classification using three CNN architectures (VGG, ResNet, and Inception). Explainable Artificial Intelligence techniques were employed to estimate the impact of segmentation. A three-classes database was composed: lung opacity (pneumonia), COVID-19, and normal. We assessed the impact of creating a CXR image database from different sources, and the COVID-19 generalization from one source to another. The segmentation achieved a Jaccard distance of 0.034 and a Dice coefficient of 0.982. The classification using segmented images achieved an F1-Score of 0.88 for the multi-class setup, and 0.83 for COVID-19 identification. In the cross-dataset scenario, we obtained an F1-Score of 0.74 and an area under the ROC curve of 0.9 for COVID-19 identification using segmented images. Experiments support the conclusion that even after segmentation, there is a strong bias introduced by underlying factors from different sources.Lucas O. TeixeiraRodolfo M. PereiraDiego BertoliniLuiz S. OliveiraLoris NanniGeorge D. C. CavalcantiYandre M. G. CostaMDPI AGarticleCOVID-19chest X-raysemantic segmentationexplainable artificial intelligenceChemical technologyTP1-1185ENSensors, Vol 21, Iss 7116, p 7116 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	COVID-19 chest X-ray semantic segmentation explainable artificial intelligence Chemical technology TP1-1185
spellingShingle	COVID-19 chest X-ray semantic segmentation explainable artificial intelligence Chemical technology TP1-1185 Lucas O. Teixeira Rodolfo M. Pereira Diego Bertolini Luiz S. Oliveira Loris Nanni George D. C. Cavalcanti Yandre M. G. Costa Impact of Lung Segmentation on the Diagnosis and Explanation of COVID-19 in Chest X-ray Images
description	COVID-19 frequently provokes pneumonia, which can be diagnosed using imaging exams. Chest X-ray (CXR) is often useful because it is cheap, fast, widespread, and uses less radiation. Here, we demonstrate the impact of lung segmentation in COVID-19 identification using CXR images and evaluate which contents of the image influenced the most. Semantic segmentation was performed using a U-Net CNN architecture, and the classification using three CNN architectures (VGG, ResNet, and Inception). Explainable Artificial Intelligence techniques were employed to estimate the impact of segmentation. A three-classes database was composed: lung opacity (pneumonia), COVID-19, and normal. We assessed the impact of creating a CXR image database from different sources, and the COVID-19 generalization from one source to another. The segmentation achieved a Jaccard distance of 0.034 and a Dice coefficient of 0.982. The classification using segmented images achieved an F1-Score of 0.88 for the multi-class setup, and 0.83 for COVID-19 identification. In the cross-dataset scenario, we obtained an F1-Score of 0.74 and an area under the ROC curve of 0.9 for COVID-19 identification using segmented images. Experiments support the conclusion that even after segmentation, there is a strong bias introduced by underlying factors from different sources.
format	article
author	Lucas O. Teixeira Rodolfo M. Pereira Diego Bertolini Luiz S. Oliveira Loris Nanni George D. C. Cavalcanti Yandre M. G. Costa
author_facet	Lucas O. Teixeira Rodolfo M. Pereira Diego Bertolini Luiz S. Oliveira Loris Nanni George D. C. Cavalcanti Yandre M. G. Costa
author_sort	Lucas O. Teixeira
title	Impact of Lung Segmentation on the Diagnosis and Explanation of COVID-19 in Chest X-ray Images
title_short	Impact of Lung Segmentation on the Diagnosis and Explanation of COVID-19 in Chest X-ray Images
title_full	Impact of Lung Segmentation on the Diagnosis and Explanation of COVID-19 in Chest X-ray Images
title_fullStr	Impact of Lung Segmentation on the Diagnosis and Explanation of COVID-19 in Chest X-ray Images
title_full_unstemmed	Impact of Lung Segmentation on the Diagnosis and Explanation of COVID-19 in Chest X-ray Images
title_sort	impact of lung segmentation on the diagnosis and explanation of covid-19 in chest x-ray images
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/a75d2b90afe9475282305ab023dd4324
work_keys_str_mv	AT lucasoteixeira impactoflungsegmentationonthediagnosisandexplanationofcovid19inchestxrayimages AT rodolfompereira impactoflungsegmentationonthediagnosisandexplanationofcovid19inchestxrayimages AT diegobertolini impactoflungsegmentationonthediagnosisandexplanationofcovid19inchestxrayimages AT luizsoliveira impactoflungsegmentationonthediagnosisandexplanationofcovid19inchestxrayimages AT lorisnanni impactoflungsegmentationonthediagnosisandexplanationofcovid19inchestxrayimages AT georgedccavalcanti impactoflungsegmentationonthediagnosisandexplanationofcovid19inchestxrayimages AT yandremgcosta impactoflungsegmentationonthediagnosisandexplanationofcovid19inchestxrayimages
_version_	1718431602997985280

Impact of Lung Segmentation on the Diagnosis and Explanation of COVID-19 in Chest X-ray Images

Ejemplares similares