CovNet-UFCSPA: assisting in the diagnosis of pneumonia by coronavirus
DOI:
https://doi.org/10.59681/2175-4411.v16.iEspecial.2024.1377Keywords:
COVID-19, Deep Learning, CNNAbstract
Objetivo: O presente estudo introduz a arquitetura CovNet-UFCSPA, que incorpora dados de pré-processamento de imagens clínicas (raio-X) e algoritmos de aprendizado profundo. Método: Utilizou-se um total de 24.235 imagens para treinamento, validação e teste do modelo, identificando áreas nos raios X que influenciam a decisão do modelo. Resultado: A arquitetura atingiu um recall de 99% na classificação de raios X de pacientes do Hospital de Clínicas de Porto Alegre (HCPA). A aplicação da técnica CLAHE melhorou a região de interesse do raio-X, reduzindo a taxa de falsos negativos de 187 para 9. Conclusão: Comparada com as arquiteturas Resnet50 V2 e Inception V3, a CovNet-UFCSPA demonstrou superioridade em taxas de falsos negativos, verdadeiros positivos e recall.
References
Gong J, Dong H, Xia SQ, Huang YZ, Wang D, Zhao Y, Liu W, Tu S, Zhang M, Wang Q, et al. Correlation analysis between disease severity and inflammation-related parameters in patients with COVID-19 pneumonia. MedRxiv. 2020. DOI: https://doi.org/10.1101/2020.02.25.20025643
Udugama B, Kadhiresan P, Kozlowski HN, Malekjahani A, Osborne M, Li VYC, Chen H, Mubareka S, Gubbay JB, Chan WCW. Diagnosing COVID-19: the disease and tools for detection. ACS Nano. 2020;14(4):3822-3835. DOI: https://doi.org/10.1021/acsnano.0c02624
DATASUS. Equipments of Imaging Used in Health - E - DATASUS. DATASUS. Available at: http://tabnet.datasus.gov.br/tabdata/LivroIDB/2edrev/e18.pdf
Chassagnon G, Vakalopoulou M, Paragios N, Revel MP. Artificial intelligence applications for thoracic imaging. Eur J Radiol. 2020;123:108774. DOI: https://doi.org/10.1016/j.ejrad.2019.108774
Nahid AA, Sikder N, Bairagi AK, Razzaque M, Masud M, Kouzani AZ, Mahmud MA, et al. A novel method to identify pneumonia through analyzing chest radiographs employing a multichannel convolutional neural network. Sensors. 2020;20(12):3482. DOI: https://doi.org/10.3390/s20123482
Rajaraman S, Antani S. Training deep learning algorithms with weakly labeled pneumonia chest X-ray data for COVID-19 detection. medRxiv. 2020. DOI: https://doi.org/10.1101/2020.05.04.20090803
Ozturk T, Talo M, Yildirim EA, Baloglu UB, Yildirim O, Acharya UR. Automated detection of COVID-19 cases using deep neural networks with X-ray images. Comput Biol Med. 2020;121:103792. DOI: https://doi.org/10.1016/j.compbiomed.2020.103792
Mittal A, Singh K, Misra DP. Detecting COVID-19 using ResNet deep learning model with X-ray images. Biocybernetics and Biomedical Engineering. 2020.
Takara, K., Nishiyama, Y., & Sone, S. (2022). Artificial Intelligence System for Chest X-ray Diagnosis of COVID-19: Development and Validation Study. Journal of Medical Internet Research, 24(1), e30527.
Nouara Cândida Xavier, Tathiane Alves Pianoschi Alva, Carla Diniz Lopes Becker. Ciências da Saúde: uma abordagem holística. Editora Conhecimento Livre; 2022. Cap 5.
Gonzalez, Rafael C., and Richard E. Woods. Processamento de imagens digitais. Editora Blucher, 2000.
Chollet F. Deep learning with Python. Simon and Schuster; 2021.
Yamashita R, Nishio M, Do RK, Togashi K. Convolutional neural networks: an overview and application in radiology. Insights into Imaging. 2018;9(4):611-629. DOI: https://doi.org/10.1007/s13244-018-0639-9
O'Shea K, Nash R. An introduction to convolutional neural networks. arXiv preprint arXiv:1511.08458. 2015.
Albawi S, Mohammed TA, Al-Zawi S. Understanding of a convolutional neural network. In: 2017 International Conference on Engineering and Technology (ICET). IEEE; 2017. pp. 1-6. DOI: https://doi.org/10.1109/ICEngTechnol.2017.8308186
scikit. Sklearn.utils.class_weight.compute_class_weight. [Online]. Available in: https://scikit-learn.org/stable/modules/generated/sklearn.utils.class_weight.compute_class_weight.html. Access at: 2024.
Cross-validation: evaluating estimator performance. [Online]. Available in: https://scikit-learn.org/stable/modules/cross_validation.html. Access at: 2024
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Submission of a paper to Journal of Health Informatics is understood to imply that it is not being considered for publication elsewhere and that the author(s) permission to publish his/her (their) article(s) in this Journal implies the exclusive authorization of the publishers to deal with all issues concerning the copyright therein. Upon the submission of an article, authors will be asked to sign a Copyright Notice. Acceptance of the agreement will ensure the widest possible dissemination of information. An e-mail will be sent to the corresponding author confirming receipt of the manuscript and acceptance of the agreement.
