DINOHist: Fine-Tuning Self-Supervised Foundation Models for Histopathological Image Retrieval

• José Solenir L. Figuerêdo Universidade Estadual de Feira de Santana (UEFS) / Instituto Federal de Sergipe (IFS)
• Rodrigo Tripodi Calumby Universidade Estadual de Feira de Santana (UEFS)

DOI: https://doi.org/10.5753/sbbd.2026.249251

Resumo

This study investigates self-supervised learning (SSL) for feature extraction in Content-Based Histopathological Image Retrieval (CBHIR). We evaluate pre-trained, domain-specific, and fine-tuned DINOv2 models (DINOHist-S and DINOHist-B) on glomerular datasets. The analysis includes comparisons with state-of-the-art histopathology-specific models to assess domain adaptation. Results demonstrate that fine-tuning enables DINOv2 to outperform state-of-the-art specialized models, even with limited labeled data. The proposed models achieve superior effectiveness, particularly at lower-ranking positions, with statistical significance. These findings show that fine-tuned foundation models can surpass specialized approaches while reducing computational costs.

Referências

Abels, E. et al. (2019). Computational pathology definitions, best practices, and recommendations for regulatory guidance: a white paper from the digital pathology association. J Pathol, 249(3):286–294.

Caron, M. et al. (2020). Unsupervised learning of visual features by contrasting cluster assignments. NIPS’20, Red Hook, NY, USA. Curran Associates Inc.

Cazzolato, M., Rodrigues, L., Scabora, L., Zaboti, G., Vasconcelos, G., Chino, D., Jorge, A., Cordeiro, R., Traina-Jr, C., and Traina, A. (2019). A dbms-based framework for content-based retrieval and analysis of skin ulcer images in medical practice. In Anais do XXXIV Simpósio Brasileiro de Banco de Dados, pages 109–120, Porto Alegre, RS, Brasil. SBC.

Chagas, P., Souza, L., Araújo, I., Aldeman, N., Duarte, A., Angelo, M., Dos-Santos, W., and Oliveira, L. (2020). Classification of glomerular hypercellularity using convolutional features and support vector machine. Artificial Intelligence in Medicine, 103:101808.

Chen, L. et al. (2019). Self-supervised learning for medical image analysis using image context restoration. Medical Image Analysis, 58:101539.

Chen, X., Xie, S., and He, K. (2021). An empirical study of training self-supervised vision transformers. In ICCV, pages 9620–9629, Los Alamitos, CA, USA. IEEE Computer Society.

Erfankhah, H. et al. (2019). Heterogeneity-aware local binary patterns for retrieval of histopathology images. IEEE Access, 7:18354–18367.

Filiot, A. et al. (2023). Scaling self-supervised learning for histopathology with masked image modeling. medRxiv.

Gudivada, V. and Raghavan, V. (1995). Content based image retrieval systems. Computer, 28(9):18–22.

Gui, J., Chen, T., Zhang, J., Cao, Q., Sun, Z., Luo, H., and Tao, D. (2024). A survey on self-supervised learning: Algorithms, applications, and future trends. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(12):9052–9071.

Haq, N. F. et al. (2021). A deep community based approach for large scale content based x-ray image retrieval. Medical Image Analysis, 68:101847.

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, pages 770–778.

Hegde, N. et al. (2019). Similar image search for histopathology: Smily. npj Digital Medicine, 2(1):56.

Hosseini, M. S., Bejnordi, B. E., Trinh, V. Q.-H., Chan, L., Hasan, D., Li, X., Yang, S., Kim, T., Zhang, H., Wu, T., Chinniah, K., Maghsoudlou, S., Zhang, R., Zhu, J., Khaki, S., Buin, A., Chaji, F., Salehi, A., Nguyen, B. N., Samaras, D., and Plataniotis, K. N. (2024). Computational pathology: A survey review and the way forward. Journal of Pathology Informatics, 15:100357.

Juan-Ferrer, P., Pérez-Sánchez, N., Pla, F., Mollineda, R. A., Roselló-Sastre, E., and Tajahuerce, E. (2026). Glomeruli detection and classification in histopathological images using deep learning semantic segmentation. BMC Medical Imaging, 26(1):103.

Kalra, S. et al. (2020). Yottixel – an image search engine for large archives of histopathology whole slide images. Medical Image Analysis, 65:101757.

Li, C. et al. (2022). Efficient self-supervised vision transformers for representation learning.

L’Imperio, V. et al. (2021). Digital pathology for the routine diagnosis of renal diseases: a standard model. Journal of Nephrology, 34(3):681–688.

Mohammad Alizadeh, S. et al. (2023). A novel siamese deep hashing model for histopathology image retrieval. Expert Systems with Applications, 225:120169.

Navid Farahani, A. et al. (2015). Whole slide imaging in pathology: advantages, limitations, and emerging perspectives. Pathology and Laboratory Medicine International, 7:23–33.

Oquab, M. et al. (2023). Dinov2: Learning robust visual features without supervision.

Shi, X. et al. (2018). Pairwise based deep ranking hashing for histopathology image classification and retrieval. Pattern Recognition, 81:14–22.

Souid, A. et al. (2023). Improving diagnosis accuracy with an intelligent image retrieval system for lung pathologies detection: a features extractor approach. Scientific Reports, 13(1):16619.

Wang, X., Du, Y., Yang, S., Zhang, J., Wang, M., Zhang, J., Yang, W., Huang, J., and Han, X. (2023). Retccl: Clustering-guided contrastive learning for whole-slide image retrieval. Medical Image Analysis, 83:102645.

Wickstrøm, K. K. et al. (2023). A clinically motivated self-supervised approach for content-based image retrieval of ct liver images. Computerized Medical Imaging and Graphics, 107:102239.

Yang, P. et al. (2020). A deep metric learning approach for histopathological image retrieval. Methods, 179:14–25. Interpretable machine learning in bioinformatics.

Zheng, Y. et al. (2022). Encoding histopathology whole slide images with location-aware graphs for diagnostically relevant regions retrieval. Medical Image Analysis, 76:102308.

Zhou, J. et al. (2022). ibot: Image bert pre-training with online tokenizer.