Classificação de Imagens Histopatológicas Usando uma VGG e Mecanismo de Atenção para Detecção de Câncer de Mama

• Marcelo Luis Rodrigues Filho IFMA
• Omar Andres Carmona Cortes IFMA

DOI: https://doi.org/10.5753/sbcas_estendido.2023.229388

Abstract

Breast cancer is a severe illness caused by uncontrolled cell division. Thus, early detection is critical to improve the patient’s life or even cure the disease. In this context, this work proposes to use a convolutional neural network, the so-called miniVGG, devised by seven layers and derived from traditional VGG architectures. Results have shown that a less deep network combined with the proper choice of learning rate and the use of attention mechanism overcomes the performance of traditional VGGs in both time and classification performance, reaching an accuracy of 93%, precision of 91%, Recall of 92%, F1 Score of 92%, and AUC of 99%.

References

Alom, M. Z., Yakopcic, C., Taha, T., and Asari, V. (2019). Breast cancer classification from histopathological images with inception recurrent residual convolutional neural network. Journal of Digital Imaging, 32.

Chaudhari, S., Polatkan, G., Ramanath, R., and Mithal, V. (2019). An Attentive Survey of Attention Models. CoRR, abs/1904.02874.

Han, Z., Wei, B., Zheng, Y., Yin, Y., Li, K., and Li, S. (2017). Breast cancer multiclassification from histopathological images with structured deep learning model. Scientific Reports, 7.

Hu, J., Shen, L., and Sun, G. (2017). Squeeze-and-excitation networks. CoRR, abs/1709.01507.

OMS (2021). Breast Cancer. Disponível em: [link].

Simonyan, K. and Zisserman, A. (2015). Very deep convolutional networks for large-scale image recognition. In International Conference on Learning Representations.

Sitaula, C. and Hossain, M. (2020). Attention-based VGG-16 model for covid-19 chest x-ray image classification. Applied Intelligence, 51:2850–2863.

Smith, L. N. (2015). Cyclical learning rates for training neural networks. CoRR, abs/1506.01186.

Smith, L. N. and Topin, N. (2017). Super-convergence: Very fast training of networks using large learning rates. CoRR, abs/1708.07120.

Spanhol, F. A., Oliveira, L. S., Petitjean, C., and Heutte, L. (2016). A Dataset for Breast Cancer Histopathological Image Classification. IEEE Transactions on Biomedical Engineering, 63(7):1455–1462.

Younis, A., Qiang, L., Nyatega, C. O., Adamu, M. J., and Kawuwa, H. B. (2022). Brain tumor analysis using deep learning and vgg-16 ensembling learning approaches. Applied Sciences, 12(14).