Reinhard it: Normalization and Classification on HER2 images

• Luan Matheus Trindade Dalmazo UFPR
• Gabriel Silva Hermida UFPR
• Sergio Ossamu Ioshii Erasto Gaertner Hospital
• Lucas Ferrari de Oliveira UFPR

DOI: https://doi.org/10.5753/sbcas_estendido.2025.7722

Resumo

HER2-positive breast cancer is one of the most aggressive subtypes and among the most frequently diagnosed. It results from the overexpression of the HER2 protein, which is assessed using the ImmunoHistoChemistry (IHC) score. However, this evaluation is often performed manually, creating opportunities for automation. In this context, this study investigates the impact of the Reinhard technique compared to mean and standard deviation normalization methods to quantify protein levels. The results demonstrate the potential of the proposed approach, achieving an F1-score of 0.89, precision of 0.89, and recall of 0.90, whereas the mean and standard deviation normalization method obtained 0.85 for precision and 0.86 for both F1-score and recall.

Referências

Assolari, C. L. and de Freitas, P. M. (2023). Automação do método de quantificação do Índice proliferativo ki-67 do câncer de mama.

Cordeiro, C. Q. (2019). An automatic patch-based approach for her-2 scoring in immuno-histochemical breast cancer images. Master’s thesis, Universidade Federal do Paraná, Curitiba, PR.

P, P. G., Senapati, K., and Pandey, A. K. (2024). A novel decision level class-wise ensemble method in deep learning for automatic multi-class classification of her2 breast cancer hematoxylin-eosin images. IEEE Access, 12:46093–46103.

Panda, S., Jangid, M., and Jain, A. (2022). Enhancing background luminance for colorectal cancer h and e stained images using modified reinhard technique. In 2022 2nd International Conference on Technological Advancements in Computational Sciences (ICTACS), pages 129–133.

Reinhard, E., Adhikhmin, M., Gooch, B., and Shirley, P. (2001). Color transfer between images. IEEE Computer Graphics and Applications, 21(5):34–41.

Thakur, V. and Kutty, R. V. (2019). Recent advances in nanotheranostics for triple negative breast cancer treatment. Journal of Experimental Clinical Cancer Research, 38(1):430.

Wang, Y., Lei, B., Elazab, A., Tan, E.-L., Wang, W., Huang, F., Gong, X., and Wang, T. (2020). Breast cancer image classification via multi-network features and dual-network orthogonal low-rank learning. IEEE Access, PP:1–1.

Yousif, M., Huang, Y., Sciallis, A., Kleer, C. G., Pang, J., Smola, B., Naik, K., McClintock, D. S., Zhao, L., Kunju, L. P., Balis, U. G. J., and Pantanowitz, L. (2021). Quantitative image analysis as an adjunct to manual scoring of er, pgr, and her2 in invasive breast carcinoma. American Journal of Clinical Pathology, 157(6):899–907.