Methodology for locating malignant thyroid nodules from infrared images

• Maira Beatriz Hernandez Moran UFF
• Aura Conci UFF

DOI: https://doi.org/10.5753/sbcas.2019.6289

Abstract

Thyroid anomalies have high prevalence and their early identifica- tion is crucial for a more effective treatment. Thermograms can be used in this process, since nodules tend to be more vascularized, resulting in a temperature increase. This work presents a methodology for determining thyroid nodules. We evaluate parameters that would allow to segment possibly nodular regions in thermographs. Convolutional neural networks (CNN) are used to classify these regions, identifying which ones refer to nodules. The good results of CNN in the classification (with 96% accuracy), show that the viability of the proposed methodology depends on the success of the segmentation.

References

Alves, M. L. D., Andrade, J., Cherri, J., Moriya, T., Piccinato, C. E., Okano, N., Llorach

Velludo, M. A., and Iazigi, N. (1988). Papel da termografia da seleção de nódulos tireoideanos de indicação cirúrgica. Arq. bras. endocrinol. metab, 32(4):97–9.

Brenner, D. J. and Hall, E. J. (2007). Computed tomography—an increasing source of radiation exposure. New England Journal of Medicine, 357(22):2277–2284.

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

Collins, A., Ring, E., Cosh, J., and Bacon, P. (1974). Quantitation of thermography in arthritis using multi-isothermal analysis. i. the thermographic index. Annals of the rheumatic diseases, 33(2):113.

Goldbach-Mansky, R., Woodburn, J., Yao, L., and Lipsky, P. E. (2003). Magnetic resonance imaging in the evaluation of bone damage in rheumatoid arthritis: a more precise image or just a more expensive one? Arthritis & Rheumatism, 48(3):585–589.

González, J., Pupo, Y., Hernandez, M., Conci, A., Machenry, T., and Fiirst, W. (2018). On image registration for study of thyroid disorders by infrared exams. In Proceedings of the International Conference on Image Processing, Computer Vision, and Pattern Recognition (IPCV), pages 151–158. The Steering Committee of The World Congress in Computer Science, Computer . . . .

Gonzáles, J. R. (2017). TUm estudo sobre a possibilidade do uso de imagens infravermelhas na análise de nódulos de tireoide. Master’s thesis, Instituto de Computação, Universidade Federal Fluminense, Niterói, RJ, Brasil.

Mahajan, P. and Madhe, S. (2014). Morphological feature extraction of thermal images for thyroid detection. Int. J. Electron. Comput. Eng, 5(4):11–14.

Moran, M., Conci, A., Gonzalez, J., Saade, D., Fiirst, W., Araújo, A., Dami˜ao, C., Lima, G., Cruz Filho, R., and Resmini, R. (2018a). Monomodal image registration by tensor analysis: going beyond the brightness constancy assumption. In 2018 IEEE/ACS 15th International Conference on Computer Systems and Applications (AICCSA), pages 1–8. IEEE.

Moran, M. B. H., Conci, A., Gonzalez, J. R., Araujo, A. S., Fiirst, W., and Damiao, C., L. G. F. R. (2018b). Identification of thyroid nodules in infrared images by convolutional neural networks. In 31st International Joint Conference on Neural Networks (IJCNN). IEEE.

Vaz, V. A. S. (2014). Diagnosis of hypo and hyperthyroid using mlpn network. International Journal of Innovative Research in Science, Engineering and Technology, 3(7):14314–23.