Classificação de Esteatose Hepática Não Alcoólica em Imagens Térmicas da Região do Fígado Utilizando Redes Neurais Convolucionais

• Daniel Moreira Pinto UFMA
• Johnatan Carvalho Souza UFMA
• Aristófanes Corrêa Silva UFMA
• Henrique Manoel de Araujo Martins Filho UAM / Alchimia
• Anselmo Cardoso de Paiva UFMA
• Renato Amaro Zângaro UAM / CITE

DOI: https://doi.org/10.5753/sbcas.2021.16074

Abstract

The non-alcoholic fatty liver disease (NAFLD) has the highest incidence among the liver diseases, afflicting about 1.5 billion people worldwide. It presents a high mortality rate and, if not dianosed and treated early, can evolve to severe liver complications. For this reason, it requires a fast and efficient diagnosis. Gauging of human body temperature with thermography is a noninvasive method of obtaining data that may indicate the presence of the disease. In this context, the purpose of this work is to present a method of NAFLD classification in thermography images of the liver. By using convolutional neural networks and image processing techniques, the proposed method achieved an accuracy of 96%, sensitivity of 91% and specificity of 97%.

References

Brioschi, M. L., Yeng, L. T., Colman, D., Silva, F. M. M., and Teixeira, M. J. (2006). Imaginologia infravermelha no estudo avançado da dor de origem visceral* infrared imaging in the advanced study of visceral pain.

Brzezinski, R. Y., Levin-Kotler, L., Rabin, N., Ovadia-Blechman, Z., Zimmer, Y., Sternfeld, A., Finchelman, J. M., Unis, R., Lewis, N., Tepper-Shaihov, O., et al. (2020). Automated thermal imaging for the detection of fatty liver disease. Scientic reports, 10(1):1–11.

Byra, M., Styczynski, G., Szmigielski, C., Kalinowski, P., Michaowski, ., Paluszkiewicz, R., Ziarkiewicz-Wróblewska, B., Zieniewicz, K., Sobieraj, P., and Nowicki, A. (2018). Transfer learning with deep convolutional neural network for liver steatosis assessment in ultrasound images. International journal of computer assisted radiology and surgery, 13(12):1895–1903.

Cotrim, H. P., Parise, E. R., Figueiredo-Mendes, C., Galizzi-Filho, J., Porta, G., and Oliveira, C. P. (2016). Nonalcoholic fatty liver disease brazilian society of hepatology consensus. Arquivos de gastroenterologia, 53(2):118–122.

Farooq, M. A. and Corcoran, P. (2020). Infrared imaging for human thermography and breast tumor classication using thermal images. In 2020 31st Irish Signals and Systems Conference (ISSC), pages 1–6. IEEE.

Géron, A. (2019). Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow: Concepts, tools, and techniques to build intelligent systems. O’Reilly Media.

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

Ozougwu, J. C. (2017). Physiology of the liver. International Journal of Research in Pharmacy and Biosciences, 4(8):13–24.

Santana, J. T., Mota, A. V. H., Gonzaga, Y. H. G., Gomes, R. M. O. P., Melo, L. C., Noronha, V. F. C. M., Santos, A. C. O. L., de Jesus, J. B., Lima, S. O., and Cruz, J. F. (2021). Perl metabólico e antropométrico dos pacientes obesos e não obesos portadores de esteatose hepática não alcoólica. Revista Eletrônica Acervo Saúde, 13(2):e5525–e5525.

Schwabe, R. F., Tabas, I., and Pajvani, U. B. (2020). Mechanisms of brosis development in nonalcoholic steatohepatitis. Gastroenterology, 158(7):1913–1928.

Shorten, C. and Khoshgoftaar, T. M. (2019). A survey on image data augmentation for deep learning. Journal of Big Data, 6(1):1–48.

Silva, L., Saade, D., Sequeiros, G., Silva, A., Paiva, A., Bravo, R., and Conci, A. (2014). A new database for breast research with infrared image. Journal of Medical Imaging and Health Informatics, 4(1):92–100.

Wahab, A. A., Salim, M., Yunus, J., and Ramlee, M. H. (2018). Comparative evaluation of medical thermal image enhancement techniques for breast cancer detection. Journal of Engineering and Technological Sciences, 50:40–52.

Zuluaga-Gomez, J., Al Masry, Z., Benaggoune, K., Meraghni, S., and Zerhouni, N. (2020). A cnn-based methodology for breast cancer diagnosis using thermal images. Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, pages 1–15.