Aplicando MultiInstance Learning (MIL) para o Diagnóstico de Câncer de Mama em Imagens Histopatológicas

• Mario Pinto Freitas UFMA
• Marcos Gabriel Mendes Lauande UFMA
• Geraldo Braz Júnior UFMA
• Marcus Vinicius Oliveira UFMA
• Gabriel Costa UFMA
• Matheus Levy UFMA
• Anselmo Cardoso de Paiva UFMA
• João D. Sousa de Almeida UFMA

DOI: https://doi.org/10.5753/sbcas.2022.222673

Abstract

Breast cancer is one of the most common cancers among women because it accounts for 29.07% of cases. Among all the variations of the disease, breast cancer is the most frequent in Brazil. And for this reason it is imperative that techniques are developed that speed up the detection process of these tumors to reduce the rate of terminal cases. Deep learning has become a strong ally for pathologists in the analysis of histopathological images making decisions quickly and reliably. In this work we present an approach based on MIL - Multi Instance Learning that has a different approach from the traditional one due to the fact that it works with several instances of the same image. We used the BreakHis breast cancer dataset to evaluate this method. In the experiments carried out, an accuracy of 90% and up to 98% in sensitivity were achieved in this binary classification problem (Benign or Malignant).

References

Alharbi, A., Wang, Y., and Zhang, Q. (2021). Trans-Attention Multiple Instance Learning for Cancer Tissue Classification in Digital Histopathology Images, page 79-84. Association for Computing Machinery, New York, NY, USA.

Cruz-Roa, A., Basavanhally, A., González, F., Gilmore, H., Feldman, M., Ganesan, S., Shih, N., Tomaszewski, J., and Madabhushi, A. (2014). Automatic detection of invasive ductal carcinoma in whole slide images with convolutional neural networks. In Gurcan, M. N. and Madabhushi, A., editors, Medical Imaging 2014: Digital Pathology, volume 9041, pages 1 - 15. International Society for Optics and Photonics, SPIE.

da Silva, A., Júnior, I., and Amora, M. (2019). Breast cancer detection in histopathological images using convolutional neural networks. In Anais do XIX Simpósio Brasileiro de Computação Aplicada à Saúde, pages 1-9, Porto Alegre, RS, Brasil. SBC.

Das, K., Conjeti, S., Chatterjee, J., and Sheet, D. (2020). Detection of breast cancer from whole slide histopathological images using deep multiple instance cnn. IEEE Access, 8:213502-213511.

Dauphin, Y. N., Fan, A., Auli, M., and Grangier, D. (2016). Language modeling with gated convolutional networks. CoRR, abs/1612.08083.

Dietterich, T. G., Lathrop, R. H., and Lozano-Pérez, T. (1997). Solving the multiple instance problem with axis-parallel rectangles. Artificial intelligence, 89(1-2):31-71.

Hering Jan, K. J. (2020). Multiple Instance Learning Via Deep Hierarchical Exploration for Histology Image Classification. International Symposium on Biomedical Imaging (ISBI), 17th edition.

Huang, G., Liu, Z., Van Der Maaten, L., and Weinberger, K. Q. (2017). Densely connected convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4700-4708.

Ilse, M., Tomczak, J., and Welling, M. (2018). Attention-based deep multiple instance learning. In International conference on machine learning, pages 2127-2136. PMLR.

INCA (2021). Types of cancer - national cancer institute josé alencar gomes da silva inca. https://www.inca.gov.br/o-que-e-cancer.

Linkon, A. H. M., Labib, M. M., Hasan, T., Hossain, M., and Marium-E-Jannat (2021). Deep learning in prostate cancer diagnosis and gleason grading in histopathology images: An extensive study. Informatics in Medicine Unlocked, 24:100582.

Patil, A., Tamboli, D., Meena, S., Anand, D., and Sethi, A. (2019). Breast cancer histopathology image classification and localization using multiple instance learning. In 2019 IEEE International WIE Conference on Electrical and Computer Engineering (WIECON-ECE), pages 1-4.

Pereira, caroline Silva, F., Pereira, L. R. M., and Couto, S. D. (2021). Procedimentos para diagnóstico do câncer de mama: Revisão de literatura. Repositório Universitário da Anima (RUNA).

Rony, J., Belharbi, S., Dolz, J., Ayed, I. B., McCaffrey, L., and Granger, E. (2019). Deep weakly-supervised learning methods for classification and localization in histology images: a survey. CoRR, abs/1909.03354.

Santos, J., Veras, R., Silva, R., Aldeman, N., Aires, K., and Bianchi, A. (2019). Classificação de imagens de biópsias renais com glomeruloesclerose segmentar e focal ou com lesóƒes mínimas utilizando transfer learning em cnn. In Anais do XIX Simpósio Brasileiro de Computação Aplicada à Saúde, pages 82-93, Porto Alegre, RS, Brasil. SBC.

Spanhol, F. A., Oliveira, L. S., Cavalin, P. R., Petitjean, C., and Heutte, L. (2017). Deep features for breast cancer histopathological image classification. In 2017 IEEE International Conference on Systems, Man, and Cybernetics (SMC), pages 1868-1873.

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.

Srinidhi, C. L., Ciga, O., and Martel, A. L. (2021). Deep neural network models for computational histopathology: A survey. Medical Image Analysis, 67:101813.

Sudharshan, P., Petitjean, C., Spanhol, F., Oliveira, L. E., Heutte, L., and Honeine, P. (2019). Multiple instance learning for histopathological breast cancer image classification. Expert Systems with Applications, 117:103-111.

Vogado, L., Veras, R., Araujo, F., Silva, R., and Aires, K. (2019). Rede neural convolucional para o diagnóstico de leucemia. In Anais do XIX Simpósio Brasileiro de Computação Aplicada à Saúde, pages 46-57, Porto Alegre, RS, Brasil. SBC.

WHO (2020). Cancer today - world health organization.

Xu, G., Song, Z., Sun, Z., Ku, C., Yang, Z., Liu, C., Wang, S., Ma, J., and Xu, W. (2019). Camel: A weakly supervised learning framework for histopathology image segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 10682-10691.

Zaheer, M., Kottur, S., Ravanbakhsh, S., Poczos, B., Salakhutdinov, R. R., and Smola, A. J. (2017). Deep sets. Advances in neural information processing systems, 30.