DSFF-GAN: A novel stain transfer network for generating immunohistochemical image of endometrial cancer

• Published in: Computers in biology and medicine Vol. 170; p. 108046
• Main Authors: Ma, Yihao, Zhang, Yiqiong, Wang, Zhengrong, Li, Juan, Miao, Yuehong, Yang, Fan, Pan, Wei
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.03.2024; Elsevier Limited
• Subjects: Breast cancer; Cancer therapies; Datasets; Deep learning; Dual-scale feature fusion block; Endometrial cancer; Endometrium; Evaluation; Generative adversarial networks; Image processing; Image quality; Immunohistochemical image; Immunohistochemistry; Localization; Medical imaging; Pathology; Similarity; Stain transfer; Staining; Uterine cancer; Virtual networks; Dual-scale feature fusion block; Immunohistochemical image; Endometrial cancer; Stain transfer
• DOI: 10.1016/j.compbiomed.2024.108046
• ISSN: 0010-4825

Immunohistochemistry (IHC) is a commonly used histological examination technique. Compared to Hematoxylin and Eosin (H&E) staining, it enables the examination of protein expression and localization in tissues, which is valuable for cancer treatment and prognosis assessment, such as the detection and diagnosis of endometrial cancer. However, IHC involves multiple staining steps, is time-consuming and expensive. One potential solution is to utilize deep learning networks to generate corresponding virtual IHC images from H&E images. However, the similarity of the IHC image generated by the existing methods needs to be further improved. In this work, we propose a novel dual-scale feature fusion (DSFF) generative adversarial network named DSFF-GAN, which comprises a cycle structure-color similarity loss, and DSFF block to constrain the model's training process and enhance its stain transfer capability. In addition, our method incorporates labeling information of positive cell regions as prior knowledge into the network to further improve the evaluation metrics. We train and test our model using endometrial cancer and publicly available breast cancer IHC datasets, and compare it with state-of-the-art methods. Compared to previous methods, our model demonstrates significant improvements in most evaluation metrics on both datasets. The research results show that our method further improves the quality of image generation and has potential value for the future clinical application of virtual IHC images. •A novel stain transfer network called DSFF-GAN was developed to generate IHC images.•A detailed image registration description for large-scale stained pathology images.•DSFF block and S–C Loss were proposed to improve network performance.•Positive region labels were input into the network to enhance the evaluation metrics.•The model was verified on both endometrial cancer and publicly breast cancer datasets.