Multi-Class Ovarian Cancer Subtype Classification in Histopathological Images Using Swin Transformer

• Juliany P. Costa Universidade Federal do Maranhão (UFMA) https://orcid.org/0009-0009-8937-073X
• Igor R. B. Estrela Universidade Federal do Maranhão (UFMA)
• Antônio O. de Carvalho Filho Universidade Federal do Piauí (UFPI)
• Anselmo C. Paiva Universidade Federal do Maranhão (UFMA)
• Aristófanes C. Silva Universidade Federal do Maranhão (UFMA)
• João O. B. Diniz Universidade Federal do Maranhão (UFMA) / Instituto Federal de Educação, Ciência e Tecnologia do Maranhão (IFMA)

DOI: https://doi.org/10.5753/bsb.2025.15159

Resumo

Ovarian cancer has a high mortality rate due to late diagnosis, which drives ongoing research to improve histopathological image analysis through artificial intelligence (AI). This study proposes a method for the automatic classification of four ovarian cancer subtypes using the Ovarian Cancer and Subtypes Histopathology Dataset, incorporating image processing techniques, data augmentation, and deep learning. The best performance was achieved with the Swin Transformer architecture, reaching an average accuracy of 98%. These results suggest that the proposed method is a promising approach for classifying ovarian cancer subtypes in histopathological images.

Palavras-chave: Histopathology, Deep Learning, Artificial Intelligence in Pathology, Cancer Subtype Classification, Medical Image Analysis

Referências

Breen, J., Allen, K., Zucker, K., Godson, L., Orsi, N., and Ravikumar, N. (2024). Histopathology foundation models enable accurate ovarian cancer subtype classification. arXiv preprint arXiv:2405.09990.

Breen, J., Allen, K., Zucker, K., Godson, L., Orsi, N. M., and Ravikumar, N. (2025). A comprehensive evaluation of histopathology foundation models for ovarian cancer subtype classification. NPJ Precision Oncology, 9(1):33.

Gonzalez, R. C. and Woods, R. E. (2018). Digital Image Processing. Pearson, 4th edition.

He, K., Zhang, X., Ren, S., and Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 770–778.

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 (CVPR), pages 4700–4708.

INCA (2023). Cancer incidence estimates in brazil – 2023 [in portuguese]. [link].

Kasture, S., Mahajan, A., Patil, P., and Deshpande, S. (2021). A New Deep Learning Method for Automatic Ovarian Cancer Prediction. International Journal of Advanced Computer Science and Applications, 12(5):560–566.

Kussaibi, H., Alibrahim, E., Alamer, E., Alhaji, G., Alshehab, S., Shabib, Z., Alsafwani, N., and Menezes, R. G. (2024). Lightgbm-based classification of ovarian cancer subtypes from histological images using resnet50. medRxiv.

Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., and Guo, B. (2021). Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 10012–10022.

Radhakrishnan, M., Sampathila, N., Muralikrishna, H., and Swathi, K. S. (2024). Advancing ovarian cancer diagnosis through deep learning and explainable ai: A multi-classification approach. IEEE Access, 12:116968–116986.

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

Vedana, A. B., Benevides, A. B., Farage, A. P. C., Moreira, G. A., Máximo, L. R. M. I., Brandão, R. S., Ubiali, I. R., Jardini, E. P., Bezerra, P. T. B., Da Silva, R. J. C., et al. (2024). Artificial intelligence in diagnostic medicine [in portuguese]. Brazilian Journal of Implantology and Health Sciences, 6(11):765–794.

Vilela, A. D., Silva, J. P., and Oliveira, M. C. (2022). Evaluation metrics for machine learning models in healthcare [in portuguese]. Revista de Engenharia e Pesquisa Aplicada, 7(2):34–45.

Werner, D. A. (2019). The fourth industrial revolution and artificial intelligence: A study on its concepts, impacts, and possible applications in law through legal text analysis for predictive classification of judicial decisions [in portuguese]. Master’s thesis, Universidade do Vale do Rio dos Sinos (UNISINOS), São Leopoldo, RS, Brazil.