Classificação de Nódulos Pulmonares Utilizando Redes Neurais Convolucionais 3D

• Thiago Lima UFPI
• Daniel Luz UFPI
• Rodrigo Veras UFPI
• Flavio Araújo UFPI

DOI: https://doi.org/10.5753/sbcas.2020.11507

Abstract

Lung cancer is the second most common in Brazil. The early detection of pulmonary nodules is essential for patient survival. In this work we propose an algorithm based on the 3D Convolutional Neural Network to classify pulmonary nodules as benign or malignant in computed tomography images. The proposed architecture has two blocks of convolutional layers followed by a pooling layer, two fully connected layers and a softmax. The use of data augmentation to balance the training set produced promising results, with accuracy of 0.9077, kappa of 0.7774, sensitivity of 0.8481, specificity of 0.9322 and AUC of 0.8901, in the classification of pulmonary nodules.

References

Armato III, S. et al. (2011). The lung image database consortium (lidc) and image database resource initiative (idri): A completed reference database of lung nodules on ct scans. Medical Physics, 38:915–931.

Avelar, A. (2019). O que e auc e roc nos modelos de machine learning. Disponível em: https://medium.com/@eam.avelar/o-que-%C3%A9-auc-e-roc-nos-modelos-demachine-learning-2e2c4112033d. Accessed: 2020-02-15.

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

de Carvalho Filho, A. O., de Sampaio, W. B., Silva, A. C., de Paiva, A. C., Nunes, R. A., and Gattass, M. (2014). Automatic detection of solitary lung nodules using quality threshold clustering, genetic algorithm and diversity index. Artificial intelligence in medicine, 60(3):165–177.

de Carvalho Filho, A. O., , et al. (2017). Computer-aided diagnosis system for lung nodules based on computed tomography using shape analysis, a genetic algorithm, and svm. Medical & Biological Engineering & Computing, 55(8):1129–1146.

Deep, G., Kaur, L., and Gupta, S. (2016). Directional local ternary quantized extrema pattern: A new descriptor for biomedical image indexing and retrieval. Engineering Science and Technology, an International Journal, 19(4):1895 – 1909.

Dey, R., Lu, Z., and Hong, Y. (2018). Diagnostic classification of lung nodules using 3d neural networks. In 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), pages 774–778.

Inca (2019). Instituto Nacional do Cancer - minist erio da saúde, cáncer de pulm ao. disponível em: https://www.inca.gov.br/tipos-de-cancer/cancer-de-pulmao. Acessado em: 02/08/2019.

Landis, J. R. and Koch, G. G. (1977). The measurement of observer agreement for categorical data. biometrics, pages 159–174.

Masood, A. et al. (2018). Computer-assisted decision support system in pulmonary cancer detection and stage classification on ct images. Journal of Biomedical Informatics, 79:117 – 128.

Mishra, A. (2018). Metrics to evaluate your machine learning algorithm.

Disponível em: https://towardsdatascience.com/metrics-to-evaluate-your-machinelearning-algorithm-f10ba6e38234. Acessado em: 15/02/2020.

Safta, W. and Frigui, H. (2018). Multiple instance learning for benign vs. malignant classification of lung nodules in ct scans. In 2018 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT), pages 490–494. IEEE.

Shaffie, Ahmed e Soliman, A. e. K. H. A. e. T. F. e. G. M. e. D. N. e. E. A. e. K. R. e. E.B. A. (2019). A novel ct-based descriptors for precise diagnosis of pulmonary nodules. In 2019 Conferencia Internacional da IEEE sobre Processamento de Imagens (ICIP) , pages 1400–1404.

Sharma, M., Bhatt, J. S., and Joshi, M. V. (2018). Early detection of lung cancer from ct images: nodule segmentation and classification using deep learning. In Tenth International Conference on Machine Vision (ICMV 2017), volume 10696, page 106960W. International Society for Optics and Photonics.

Shorten, C. and Khoshgoftaar, T. M. (2019). A survey on image data augmentation for deep learning. Journal of Big Data, 6(1):60. https://doi.org/10.1186/ s40537-019-0197-0.

Yan, X., Pang, J., Qi, H., Zhu, Y., Bai, C., Geng, X., Liu, M., Terzopoulos, D., and Ding, X. (2016). Classification of lung nodule malignancy risk on computed tomography images using convolutional neural network: A comparison between 2d and 3d strategies. In Asian Conference on Computer Vision, pages 91–101. Springer.

Zhao, X., Liu, L., Qi, S., Teng, Y., Li, J., and Qian, W. (2018). Agile convolutional neural network for pulmonary nodule classification using ct images. International journal of computer assisted radiology and surgery, 13(4):585–595.