Segmentação Multi-classes em termografias mamárias utilizando Redes Profundas
Abstract
Despite the many advances in medicine and science to combat breast cancer, recent studies regarding the incidence of the disease in Brazil and in the world show that this disease is one of the main causes of death among women. In order to collaborate with the early diagnosis of breast anomalies, increasing the chances of cure, breast thermography image has been used with the purpose of collaborating where mammography is unfavorable. Therefore, this work proposes to use the static protocol of thermography since it offers 5 points of view of the breast region, thus expanding the possibility of finding carcinomes. And, to find the region of interest in these images, we used the U-Net network architecture to perform a multi-class segmentation with the aim of providing a future step of more efficient thermographic pattern recognition between healthy and abnormal breasts. Preliminary results achieved an accuracy of 80,71% and a value of 0,82% for the IoU metric.
References
FEMAMA, "Câncer de mama: entenda as diferenças entre prevenção e detecção precoce," 2019.
T. B. Borchartt, A. Conci, R. C. Lima, R. Resmini, and A. Sanchez, "Breast thermography from an image processing viewpoint: A survey," Signal Processing, vol. 93, no. 10, pp. 2785-2803, 2013, signal and Image Processing Techniques for Detection of Breast Diseases.
J. Singh and A. S. Arora, "Automated approaches for rois extraction in medical thermography: A review and future directions," Multimedia Tools Appl., vol. 79, no. 21-22, p. 15273-15296, 2020.
R. Roslidar, A. Rahman, R. Muharar, M. R. Syahputra, F. Arnia, M. Syukri, B. Pradhan, and K. Munadi, "A review on recent progress in thermal imaging and deep learning approaches for breast cancer detection," IEEE Access, vol. 8, pp. 116 176-116 194, 2020.
E. A. Mohamed, E. A. Rashed, T. Gaber, and O. Karam, "Deep learning model for fully automated breast cancer detection system from thermograms," PLOS ONE, vol. 17, no. 1, pp. 1-20, 01 2022.
L. F. da Silva, D. C. M. Saade, G. O. Sequeiros, A. C. Silva, A. C. de Paiva, R. de Souza Bravo, and A. Conci, "A new database for breast research with infrared image," Journal of Medical Imaging and Health Informatics, vol. 4, pp. 92-100, 2014.
E. Gerasimova, O. Plekhov, Y. Bayandin, O. Naimark, and G. Freynd, "Identification of breast cancer using analysis of thermal signals by nonlinear dynamics methods," 01 2012.
K. McPherson, C. M. Steel, and J. M. Dixon, "Breast cancer-epidemiology, risk factors, and genetics," BMJ, vol. 321, no. 7261, pp. 624-628, 2000.
C. N. S. Belfort, "Análise e classificação de termografias dinâmicas da mama utilizando a variação de temperatura e máquina de vetores de suporte," 2016.
E. d. L. Chaves, "Detecção de câncer de mama por meio de imagens infravermelhas utilizando redes neurais convolucionais," 2019.
L. F. d. Silva, "Uma análise híbrida para detecção de anomalias da mama usando séries temporais de temperatura," Ph.D. dissertation, Universidade Federal Fluminense, 2015.
G. O. Sequeiros, "Desenvolvimento de um banco de dados de imagens mastológicas acessíveis via web com recuperação de imagens baseada em conteúdo," Master's thesis, Universidade Federal Fluminense, 2013.
F. J. P. Andrade, A. C. Paiva, and A. C. Silva, "Análise de imagens de termografia dinâmica para classificação de alterações na mama usando séries temporais," in SIBGRAPI. Niterói, RJ, Brazil: SBC, 10 2017.
R. d. S. Marques, "Segmentação automática das mamas em imagens térmicas," Master's thesis, Universidade Federal Fluminense, 2012.
C. B. Gonçalves, "Detecção de câncer de mama utilizando imagens termográficas," 2017.
O. Ronneberger, P. Fischer, and T. Brox, "U-net: Convolutional networks for biomedical image segmentation," in Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015. Springer International Publishing, 2015, pp. 234-241.
R. Ribani and M. Marengoni, "A survey of transfer learning for convolutional neural networks," 10 2019, pp. 47-57.
H. Rezatofighi, N. Tsoi, J. Gwak, A. Sadeghian, I. Reid, and S. Savarese, "Generalized intersection over union," June 2019.
