About | Contact Us | Register | Login
ProceedingsSeriesJournalsSearchEAI
bebi 21(2): e2

Research Article

Texture-based Feature Extraction for COVID-19 Pneumonia Classification using Chest Radiography

Download1348 downloads
Cite
BibTeX Plain Text
  • @ARTICLE{10.4108/eai.4-3-2021.168864,
        author={L. V. Moura and C. M. Dartora and C. Mattjie and R. C. Barros and A. M. Marques da Silva},
        title={Texture-based Feature Extraction for COVID-19 Pneumonia Classification using Chest Radiography},
        journal={EAI Endorsed Transactions on Bioengineering and Bioinformatics},
        volume={1},
        number={2},
        publisher={EAI},
        journal_a={BEBI},
        year={2021},
        month={3},
        keywords={coronavirus, X-rays, atypical pneumonia, radiomics, texture},
        doi={10.4108/eai.4-3-2021.168864}
    }
    
  • L. V. Moura
    C. M. Dartora
    C. Mattjie
    R. C. Barros
    A. M. Marques da Silva
    Year: 2021
    Texture-based Feature Extraction for COVID-19 Pneumonia Classification using Chest Radiography
    BEBI
    EAI
    DOI: 10.4108/eai.4-3-2021.168864
L. V. Moura1, C. M. Dartora1,2, C. Mattjie1,2, R. C. Barros3, A. M. Marques da Silva1,*
  • 1: PUCRS, School of Technology, Medical Imaging Computing Laboratory – MEDICOM, Porto Alegre, Brazil
  • 2: PUCRS, School of Medicine, Graduate Program in Biomedical Gerontology, Porto Alegre, Brazil
  • 3: PUCRS, School of Technology, Machine Learning Theory and Applications Lab – MALTA, Porto Alegre, Brazil
*Contact email: ana.marques@pucrs.br

Abstract

INTRODUCTION: The identification of COVID-19 pneumonia using chest radiography is challenging.

OBJECTIVES: We investigate classification models to differentiate COVID-19-based and typical pneumonia in chest radiography.

METHODS: We use 136 segmented chest X-rays to train and evaluate the performance of support vector machine (SVM), random forest (RF), AdaBoost (AB), and logistic regression (LR) classification methods. We use the PyRadiomics to extract statistical texture-based features in the right, left, and in six lung zones. We use a stratified k-folds (k=5) cross-validation within the training dataset, selecting the most relevant features with validation accuracy and relative feature importance.

RESULTS: The AB model seems to be the best discriminant method, using six lung zones (AUC = 0.98).

CONCLUSION: Our study shows a predominance of radiomic texture-based features related to COVID-19 pneumonia within the right lung, with a tendency within the upper lung zone.

Keywords
coronavirus, X-rays, atypical pneumonia, radiomics, texture
Received
2020-12-10
Accepted
2021-02-17
Published
2021-03-04
Publisher
EAI
http://dx.doi.org/10.4108/eai.4-3-2021.168864

Copyright © 2021 L. V. Moura et al., licensed to EAI. This is an open access article distributed under the terms of the Creative Commons Attribution license, which permits unlimited use, distribution and reproduction in any medium so long as the original work is properly cited.

EBSCOProQuestDBLPDOAJPortico
EAI Logo

About EAI

  • Who We Are
  • Leadership
  • Research Areas
  • Partners
  • Media Center

Community

  • Membership
  • Conference
  • Recognition
  • Sponsor Us

Publish with EAI

  • Publishing
  • Journals
  • Proceedings
  • Books
  • EUDL