Detecção de Nódulos da Tireoide em Exames de Termografia utilizando Redes Neurais Convolucionais em Cascata

• Ricardo José Fernandes Anchieta Júnior UFMA
• Italo Francyles Santos da Silva UFMA
• Aristófanes Corrêa Silva UFMA
• Anselmo Cardoso de Paiva UFMA

DOI: https://doi.org/10.5753/sbcas.2021.16071

Abstract

Thermography exams can be used as a non-invasive way for detecting diseases through the analysis of thermal images of the human body. The present work presents a method to detect nodules of the thyroid gland in thermal images using two cascade Convolutional Neural Networks (CNNs). The first CNN operates in the generation of candidate nodules, and the second CNN refines the detection, eliminating false positives. The experiments were carried out with thermal images of 20 patients. The final result reaches 97% accuracy, 95% recall and 80% f1-score using two AlexNet CNNs.

References

Bahramian, F. and Mojra, A. (2020). Thyroid cancer estimation using infrared thermography data. Infrared Physics & Technology, 104:103126.

Brioschi, M., Teixeira, M., Silva, F., and Colman, D. (2010). Medical thermography textbook: principles and applications. Editora e Livrarai Andreoli.

Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., and Thrun, S. (2017). Dermatologist-level classification of skin cancer with deep neural networks. nature, 542(7639):115–118.

Fiirst, W., Montenegro, A., Machenry, T., Li, H., Du, C., and Conci, A. (2018). Introducing tensor analysis for biomedical image registration. 6o. Encontro Nacional de Engenharia Biomecanica (ENEBI 2018), Aguas de Lindoia, SP (proc. ABCM, SBB).

Fiirst, W. G., Resmini, R., and Conci, A. (2019). Desenvolvimento e comparação de métodos para geração de séries temporais para imagens térmicas da tireoide. In Anais Estendidos do XIX Simpósio Brasileiro de Computação Aplicada à Saúde, pages 43– 48. SBC.

Gonzalez, J. (2017). Um estudo sobre a possibilidade do uso de imagens infravermelhas na analise de nodulos de tireoide. Dissertation, IC, Universidade Federal Fluminense.

González, J., Conci, A., Moran, M., Araujo, A., Paes, A., Damião, C., and Fiirst, W. (2019). Analysis of static and dynamic infrared images for thyroid nodules investigation. In 2019 IEEE/ACS 16th International Conference on Computer Systems and Applications (AICCSA), pages 1–7. IEEE.

Guizar-Sicairos, M., Thurman, S., and Fienup, J. (2008). Efficient subpixel image registration algorithms. Optics letters, 33 2:156–8.

He, K., Zhang, X., Ren, S., and Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778.

Hernandez Moran, M., Conci, A., and Araújo, A. (2019). Evaluation of quantitative features and convolutional neural networks for nodule identification in thyroid thermographies. pages 747–751.

Krizhevsky, A., Sutskever, I., and Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25:1097–1105.

Lévy, D. and Jain, A. (2016). Breast mass classification from mammograms using deep convolutional neural networks. arXiv preprint arXiv:1612.00542.

Lin, T.-Y., Goyal, P., Girshick, R., He, K., and Dollár, P. (2018). Focal loss for dense object detection.

McPhail, S., Johnson, S., Greenberg, D., Peake, M., and Rous, B. (2015). Stage at diagnosis and early mortality from cancer in england. British journal of cancer, 112(1):S108– S115.

Moran, M. B., Conci, A., González, J. R., Araújo, A. S., Fiirst, W., Damião, C. P., Lima, G. A., and da Cruz Filho, R. A. (2018). Identification of thyroid nodules in infrared images by convolutional neural networks. In 2018 International joint conference on neural networks (IJCNN), pages 1–7. IEEE.

Nguyen, Q. T., Lee, E. J., Huang, M. G., Park, Y. I., Khullar, A., and Plodkowski, R. A. (2015). Diagnosis and treatment of patients with thyroid cancer. American health & drug benefits, 8(1):30.

Rahib, L., Smith, B. D., Aizenberg, R., Rosenzweig, A. B., Fleshman, J. M., and Matrisian, L. M. (2014). Projecting cancer incidence and deaths to 2030: the unexpected burden of thyroid, liver, and pancreas cancers in the united states. Cancer research, 74(11):2913–2921.

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

Tan, M. and Le, Q. V. (2019). Efficientnet: Rethinking model scaling for convolutional neural networks. arXiv preprint arXiv:1905.11946.

Zhu, R., Zhang, R., and Xue, D. (2015). Lesion detection of endoscopy images based on convolutional neural network features. In 2015 8th International Congress on Image and Signal Processing (CISP), pages 372–376. IEEE.