Análise da Segmentação e Extração de Características na Detecção de COVID-19 em imagens de Raio-x de Tórax

• Camila Catiely de Sá Almondes UFPI
• Flávio Henrique Duarte de Araújo UFPI

DOI: https://doi.org/10.5753/enucompi.2021.17747

Abstract

COVID-19 mainly affects the lungs, causing shortness of breath, coughing and even multiple organ failure, making people seriously ill. A chest X-ray is necessary for testing and to assess the lung and the progress of the virus in terms of its effects. This work presents the evaluation of the descriptors Dense Net201, VGG16, RESNET50 and Xception, and the classifiers Multi-layer Perceptron (MLP) and Random Forest (RF), using the COVID-19 chest x-ray database for COVID-19 diagnosis. To evaluate the segmentation, the base Tuberculosis (TB) Chest X-ray Database was used. The tests were performed on a set of segmented images containing 6012 images of pulmonary infections, 3616 of COVID-19 and 10192 without findings. The evaluated scenarios were (Covid x Normal), (Covid x Lung Opacity), (Covid x Lung Opacity x Normal). The best results were achieved with the descriptor DenseNet201 and the MLP classifier in the scenario (Covid x Normal), with Accuracy and Kappa of 0.99.

References

Afshar, P., Heidarian, S., Naderkhani, F., Oikonomou, A., Plataniotis, K. N., and Mohammadi, A. (2020). Covid-caps: A capsule network-based framework for identication of covid-19 cases from x-ray images. arXiv preprint arXiv:2004.02696.

Chollet, F. (2017). Xception: Deep learning with depthwise separable convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 1251–1258.

da Nobrega, R. V. M. et al. (2018). Lung nodule classication via deep transfer learning in ct lung images. In 2018 IEEE 31st International Symposium on Computer-Based Medical Systems (CBMS), pages 244–249.

He, K., Zhang, X., Ren, S., and Sun, J. (2016). Deep residual learning for image recognition. In Conference on Computer Vision and Pattern Recognition, volume 2016December, pages 770–778. IEEE Computer Society.

Huang, G., Liu, Z., Van Der Maaten, L., and Weinberger, K. Q. (2017). Densely connected convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4700–4708.

Narin, A., Kaya, C., and Pamuk, Z. (2020). Automatic detection of coronavirus disease (covid-19) using x-ray images and deep convolutional neural networks. arXiv preprint arXiv:2003.10849.

Nicolas, P. R. (2015). Scala for machine learning. Packt Publishing Ltd.

Rahman, T., Khandakar, A., Kadir, M. A., Islam, K. R., Islam, K. F., Mazhar, R., Hamid, T., Islam, M. T., Kashem, S., Mahbub, Z. B., et al. (2020). Reliable tuberculosis detection using chest x-ray with deep learning, segmentation and visualization. IEEE Access, 8:191586–191601.

Simonyan, K. and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.

Suthaharan, S. (2016). Machine learning models and algorithms for big data classication. Integr. Ser. Inf. Syst, 36:1–12.

T. R. Muhammad, E. H. Chowdhury, A. K. R. M. M. A. K. Z. B. M. e. a. (2020). Covid-19 chest x-ray database. online. Available: https://www.kaggle.com/tawsifurrahman/covid19-radiography-database.

Tang, H., Kim, D. R., and Xie, X. (2018). Automated pulmonary nodule detection using 3d deep convolutional neural networks. In 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), pages 523–526.

Teixeira, L. O., Pereira, R. M., Bertolini, D., Oliveira, L. S., Nanni, L., Cavalcanti, G. D., and Costa, Y. M. (2020). Impact of lung segmentation on the diagnosis and explanation of covid-19 in chest x-ray images. arXiv preprint arXiv:2009.09780.

Varela-Santos, S. and Melin, P. (2021). A new approach for classifying coronavirus covid-19 based on its manifestation on chest x-rays using texture features and neural networks. Information sciences, 545:403–414.