Classificação Automática de Tecidos do Microambiente Tumoral no Câncer Gástrico com Ensemble por Stacking
Resumo
O câncer gástrico permanece como uma importante causa de mortalidade relacionada ao câncer, e a caracterização do Microambiente Tumoral (Tumor Microenvironment – TME) é essencial para o prognóstico e o planejamento terapêutico. No entanto, a heterogeneidade morfológica dos tecidos gástricos ainda impõe desafios à classificação automatizada, mesmo diante dos avanços em Deep Learning (DL). Este estudo propõe um ensemble por empilhamento (stacking) de modelos de DL para a classificação multiclasse de tecidos do TME em imagens histopatológicas H&E, utilizando o dataset HMU-GC-HE-30K. Oito arquiteturas pré-treinadas foram empregadas como modelos base e suas predições foram combinadas por um meta-modelo MLP, capaz de aprimorar as fronteiras de decisãoao aprender uma combinação não linear das predições dos modelos base. Técnicas de aumento de dados e aprendizado por transferência foram adotadas para mitigar limitações do conjunto de dados e aumentar a capacidade de generalização. Os resultados indicam que o stacking ajustado superou estatisticamente o baseline alcançando 81,3% de F1-score, superando os modelos individuais e sugerindo maior robustez na distinção de classes teciduais complexas. Assim, o método demonstra potencial para apoiar a decisão em patologia digital.
Referências
de Visser, K. E. and Joyce, J. A. (2023). The evolving tumor microenvironment: From cancer initiation to metastatic outgrowth. Cancer Cell, 41(3):374–403.
Demšar, J. (2006). Statistical comparisons of classifiers over multiple data sets. Journal of Machine learning research, 7(Jan):1–30.
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., and Fei-Fei, L. (2009). Imagenet: A large-scale hierarchical image database. In 2009 IEEE Conference on Computer Vision and Pattern Recognition. IEEE.
Dietterich, T. G. (2000). Ensemble methods in machine learning. In Multiple Classifier Systems, pages 1–15. Springer Berlin Heidelberg, Berlin, Heidelberg.
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., and Houlsby, N. (2021). An image is worth 16×16 words: Transformers for image recognition at scale. In International Conference on Learning Representations (ICLR), Austria. Disponível em: [link].
Ferlay, J., Ervik, M., Lam, F., Laversanne, M., Colombet, M., Mery, L., Piñeros, M., Znaor, A., Soerjomataram, I., and Bray, F. (2024). Global cancer observatory: Cancer today. Site institucional da International Agency for Research on Cancer.
Guehria, S., Belleili, H., and Azizi, N. (2023). A survey on ensemble multi-label classifiers. In Abraham, A. et al., editors, Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition (SoCPaR 2022), volume 648 of Lecture Notes in Networks and Systems, pages 100–109. Springer Nature Switzerland, Cham.
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.
Howard, A., Sandler, M., Chu, G., Chen, L.-C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., Vasudevan, V., Le, Q. V., and Adam, H. (2019). Searching for mobilenetv3. arXiv preprint arXiv:1905.02244.
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.
Iglovikov, V., Druzhinin, M., Buslaev, A., et al. (2025). Albumentations documentation. [link]. Acesso em: 12 ago. 2025.
Kablan, R., Miller, H. A., Suliman, S., and Frieboes, H. B. (2023). Evaluation of stacked ensemble model performance to predict clinical outcomes: A covid-19 study. International Journal of Medical Informatics, 175:105090.
Kather, J. N., Krisam, J., Charoentong, P., Luedde, T., Herpel, E., Weis, C.-A., and et al. (2019). Predicting survival from colorectal cancer histology slides using deep learning: A retrospective multicenter study. PLoS Medicine, 16(1):e1002730.
Lou, S., Ji, J., Li, H., and et al. (2025). A large histological images dataset of gastric cancer with tumour microenvironment annotation for ai. Scientific Data, 12:138.
Lou, S., Ji, J., Li, H., Zhang, X., Jiang, Y., Hua, M., Chen, K., Ge, K., Zhang, Q., Wang, L., Han, P., and Cao, L. (2024). Gastric cancer histopathology tissue image dataset (gchtid). Dataset.
Mandal, S., Baker, A.-M., Graham, T. A., and Bräutigam, K. (2025). The tumour histopathology “glossary” for ai developers. PLOS Computational Biology, 21(1):e1012708.
Mikhailov, I., Khvostikov, A., and Krylov, A. (2022). Methodical approaches to annotation and labeling of histological images in order to automatically detect the layers of the stomach wall and the depth of invasion of gastric cancer. Archive of Pathology = Arkhiv patologii, 84(6):67–73. In Russian.
Mohammed, M., Mwambi, H., Mboya, I. B., Elbashir, M. K., and Omolo, B. (2021). A stacking ensemble deep learning approach to cancer type classification based on tcga data. Scientific Reports, 11(1):15626.
Mumuni, A. and Mumuni, F. (2022). Data augmentation: A comprehensive survey of modern approaches. Array, 16:100258.
Naimi, A. I. and Balzer, L. B. (2018). Stacked generalization: An introduction to super learning. European Journal of Epidemiology, 33(5):459–464.
Rong, R., Sheng, H., Jin, K. W., Wu, F., Luo, D., Wen, Z., Tang, C., Yang, D. M., Jia, L., Amgad, M., Cooper, L. A. D., Xie, Y., Zhan, X., Wang, S., and Xiao, G. (2022). A deep learning approach for histology-based nuclei segmentation and tumor microenvironment characterization. bioRxiv. Preprint.
Su, X., Mao, Q., Wu, Z., et al. (2025). Large language models driven neural architecture search for universal and lightweight disease diagnosis on histopathology slide images. npj Digital Medicine, 8:682.
Taifa, I. A., Setu, D. M., Islam, T., Dey, S. K., and Rahman, T. (2024). A hybrid approach with customized machine learning classifiers and multiple feature extractors for enhancing diabetic retinopathy detection. Healthcare Analytics, 5:100346.
Tan, M. and Le, Q. V. (2019). Efficientnet: Rethinking model scaling for convolutional neural networks. In Proceedings of the 36th International Conference on Machine Learning (ICML). Disponível em: [link].
Wang, Z., Peng, H., Wan, J., and et al. (2024). Identification of histopathological classification and establishment of prognostic indicators of gastric adenocarcinoma based on deep learning algorithm. Medical Molecular Morphology, 57:286–298.
Demšar, J. (2006). Statistical comparisons of classifiers over multiple data sets. Journal of Machine learning research, 7(Jan):1–30.
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., and Fei-Fei, L. (2009). Imagenet: A large-scale hierarchical image database. In 2009 IEEE Conference on Computer Vision and Pattern Recognition. IEEE.
Dietterich, T. G. (2000). Ensemble methods in machine learning. In Multiple Classifier Systems, pages 1–15. Springer Berlin Heidelberg, Berlin, Heidelberg.
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., and Houlsby, N. (2021). An image is worth 16×16 words: Transformers for image recognition at scale. In International Conference on Learning Representations (ICLR), Austria. Disponível em: [link].
Ferlay, J., Ervik, M., Lam, F., Laversanne, M., Colombet, M., Mery, L., Piñeros, M., Znaor, A., Soerjomataram, I., and Bray, F. (2024). Global cancer observatory: Cancer today. Site institucional da International Agency for Research on Cancer.
Guehria, S., Belleili, H., and Azizi, N. (2023). A survey on ensemble multi-label classifiers. In Abraham, A. et al., editors, Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition (SoCPaR 2022), volume 648 of Lecture Notes in Networks and Systems, pages 100–109. Springer Nature Switzerland, Cham.
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.
Howard, A., Sandler, M., Chu, G., Chen, L.-C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., Vasudevan, V., Le, Q. V., and Adam, H. (2019). Searching for mobilenetv3. arXiv preprint arXiv:1905.02244.
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.
Iglovikov, V., Druzhinin, M., Buslaev, A., et al. (2025). Albumentations documentation. [link]. Acesso em: 12 ago. 2025.
Kablan, R., Miller, H. A., Suliman, S., and Frieboes, H. B. (2023). Evaluation of stacked ensemble model performance to predict clinical outcomes: A covid-19 study. International Journal of Medical Informatics, 175:105090.
Kather, J. N., Krisam, J., Charoentong, P., Luedde, T., Herpel, E., Weis, C.-A., and et al. (2019). Predicting survival from colorectal cancer histology slides using deep learning: A retrospective multicenter study. PLoS Medicine, 16(1):e1002730.
Lou, S., Ji, J., Li, H., and et al. (2025). A large histological images dataset of gastric cancer with tumour microenvironment annotation for ai. Scientific Data, 12:138.
Lou, S., Ji, J., Li, H., Zhang, X., Jiang, Y., Hua, M., Chen, K., Ge, K., Zhang, Q., Wang, L., Han, P., and Cao, L. (2024). Gastric cancer histopathology tissue image dataset (gchtid). Dataset.
Mandal, S., Baker, A.-M., Graham, T. A., and Bräutigam, K. (2025). The tumour histopathology “glossary” for ai developers. PLOS Computational Biology, 21(1):e1012708.
Mikhailov, I., Khvostikov, A., and Krylov, A. (2022). Methodical approaches to annotation and labeling of histological images in order to automatically detect the layers of the stomach wall and the depth of invasion of gastric cancer. Archive of Pathology = Arkhiv patologii, 84(6):67–73. In Russian.
Mohammed, M., Mwambi, H., Mboya, I. B., Elbashir, M. K., and Omolo, B. (2021). A stacking ensemble deep learning approach to cancer type classification based on tcga data. Scientific Reports, 11(1):15626.
Mumuni, A. and Mumuni, F. (2022). Data augmentation: A comprehensive survey of modern approaches. Array, 16:100258.
Naimi, A. I. and Balzer, L. B. (2018). Stacked generalization: An introduction to super learning. European Journal of Epidemiology, 33(5):459–464.
Rong, R., Sheng, H., Jin, K. W., Wu, F., Luo, D., Wen, Z., Tang, C., Yang, D. M., Jia, L., Amgad, M., Cooper, L. A. D., Xie, Y., Zhan, X., Wang, S., and Xiao, G. (2022). A deep learning approach for histology-based nuclei segmentation and tumor microenvironment characterization. bioRxiv. Preprint.
Su, X., Mao, Q., Wu, Z., et al. (2025). Large language models driven neural architecture search for universal and lightweight disease diagnosis on histopathology slide images. npj Digital Medicine, 8:682.
Taifa, I. A., Setu, D. M., Islam, T., Dey, S. K., and Rahman, T. (2024). A hybrid approach with customized machine learning classifiers and multiple feature extractors for enhancing diabetic retinopathy detection. Healthcare Analytics, 5:100346.
Tan, M. and Le, Q. V. (2019). Efficientnet: Rethinking model scaling for convolutional neural networks. In Proceedings of the 36th International Conference on Machine Learning (ICML). Disponível em: [link].
Wang, Z., Peng, H., Wan, J., and et al. (2024). Identification of histopathological classification and establishment of prognostic indicators of gastric adenocarcinoma based on deep learning algorithm. Medical Molecular Morphology, 57:286–298.
Publicado
01/06/2026
Como Citar
BARBOSA, François F. R.; SILVA, Ivan S.; VERAS, Rodrigo M. S.; BORGES, Rodrigo N.; RAMOS, Isaac S. S.; RECAMONDE-MENDOZA, Mariana.
Classificação Automática de Tecidos do Microambiente Tumoral no Câncer Gástrico com Ensemble por Stacking. In: SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO APLICADA À SAÚDE (SBCAS), 26. , 2026, Ouro Preto/MG.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 942-953.
ISSN 2763-8952.
DOI: https://doi.org/10.5753/sbcas.2026.21590.
