MHCAF-Net: Multi-Scale Cross-Attention Fusion for Histological Grading of Invasive Ductal Carcinoma

• Mario Vitor Vieira Cella UFMA
• Alejandro Costa de Oliveira UFMA
• Celso Luiz Silva Soares Filho UFMA
• Darlan Bruno Pontes Quintanilha UFMA
• Tiago Bonini Borchartt UFMA
• Francisco Glaubos Nunes Clímaco UFMA
• João Dallyson Sousa de Almeida Almeida UFMA

DOI: https://doi.org/10.5753/sbcas.2026.21032

Resumo

Breast cancer remains a critical global health challenge, requiring precise diagnostic tools for effective treatment planning. While mammography is the standard screening tool, histopathological analysis remains the gold standard for confirming diagnoses and detailing tissue characteristics. However, determining the histological grade of Invasive Ductal Carcinoma (IDC) presents significant morphological ambiguity, where global tissue structures often contradict local cellular details. This work introduces a architecture using Convolutional Neural Networks (CNNs) to classify the histological grade of IDC. The primary contribution resides in an attention-based fusion mechanism that integrates multi-scale representations of histopathological images. The approach achieved results of 85.29%±3.0% accuracy, 85.22%±2.9% F1-Score, 89.56%±2.7% precision, and 81.36%±4.2% recall. These results demonstrate the viability of the proposed approach, which effectively captures both tissue architecture and cellular morphology to assist pathologists in accurate medical image analysis.

Palavras-chave: Breast Cancer, Invasive Ductal Carcinoma, Histopathological Images, Convolutional Neural Networks, Multi-scale Fusion

Referências

S Agarwal, KV Arya, and YK Meena. Multifusionnet: Multilayer multimodal fusion of deep neural networks for chest x-ray image classification. arxiv. arXiv preprint arXiv:2401.00728, 2024.

Islam Alzoubi, Bowen Xin, Rolf Bjerkvig, Jian Wang, and Xiuying Wang. An adaptive multi-graph fusion for tumor grading in pathology images. Pattern Recognition, 171: 112214, 2026. ISSN 0031-3203. DOI: 10.1016/j.patcog.2025.112214.

James Bergstra, Rémi Bardenet, Yoshua Bengio, and Balázs Kégl. Algorithms for hyperparameter optimization. Advances in neural information processing systems, 24, 2011.

Hamidreza Bolhasani, Elham Amjadi, Maryam Tabatabaeian, and Somayyeh Jafarali Jassbi. A histopathological image dataset for grading breast invasive ductal carcinomas. Informatics in Medicine Unlocked, 19:100341, 2020. DOI: 10.1016/j.imu.2020.100341.

Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016.

Gao Huang, Zhuang Liu, Laurens Van Der Maaten, and Kilian Q Weinberger. Densely connected convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4700–4708, 2017.

Instituto Nacional de Câncer José Alencar Gomes da Silva INCA. Estimativa 2023: incidência de câncer no brasil. INCA, 2023.

Daisuke Komura and Shumpei Ishikawa. Machine learning methods for histopathological image analysis. Computational and structural biotechnology journal, 16:34–42, 2018.

Eelandula Kumaraswamy, Sumit Kumar, and Manoj Sharma. An invasive ductal carcinomas breast cancer grade classification using an ensemble of convolutional neural networks. Diagnostics, 13(11):1977, 2023. DOI: 10.3390/diagnostics13111977.

Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, and Baining Guo. Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF international conference on computer vision, pages 10012–10022, 2021.

Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, and Saining Xie. A convnet for the 2020s. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11976–11986, 2022.

Shallu Sharma, Sumit Kumar, Manoj Sharma, and Ashish Kalkal. An ensemble of deep cnns for automatic grading of breast cancer in digital pathology images. Neural Computing and Applications, 36(11):5673–5693, 2024. ISSN 1433-3058. DOI: 10.1007/s00521-023-09368-1.

Karen Simonyan and Andrew Zisserman. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014.

Radhakrishnan Sujatha, Jyotir Moy Chatterjee, Anastassia Angelopoulou, Epaminondas Kapetanios, Parvathaneni Naga Srinivasu, and Duraisamy Jude Hemanth. A transfer learning–based system for grading breast invasive ductal carcinoma. IET Image Processing, 17(7):1979–1990, 2023. DOI: 10.1049/ipr2.12660.

Hyuna Sung, Jacques Ferlay, Rebecca L Siegel, Mathieu Laversanne, Isabelle Soerjomataram, Ahmedin Jemal, and Freddie Bray. Global cancer statistics 2020: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: a cancer journal for clinicians, 71(3):209–249, 2021.

Mingxing Tan and Quoc Le. Efficientnet: Rethinking model scaling for convolutional neural networks. In International conference on machine learning, pages 6105–6114. PMLR, 2019.

Mitko Veta, Yujing J Heng, Nikolas Stathonikos, Babak Ehteshami Bejnordi, Francisco Beca, Thomas Wollmann, Karl Rohr, Manan A Shah, Dayong Wang, Mikael Rousson, et al. Predicting breast tumor proliferation from whole-slide images: the tupac16 challenge. Medical image analysis, 54:111–121, 2019.