Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a semi-supervised virtual staining framework that effectively utilizes both limited paired and abundant unpaired histopathological images to generate target-stained images. By incorporating Hessian-derived morphology preservation and histopathological realism constraints, the method ensures that the generated outputs maintain tissue morphology and realistic staining characteristics, addressing the challenges of training collapse and unrealistic results. Extensive experiments show significant improvements in image quality and diagnostic performance, particularly in H&E to IHC translation for Ki67 and HER2 markers.
Unpaired data can be harnessed to enhance virtual staining accuracy without the need for extensive paired datasets, revolutionizing histopathological image generation.
Virtual staining aims to computationally generate target-stained histopathological images while reducing the cost and time associated with conventional staining procedures. However, existing methods rely predominantly on strictly paired and accurately registered training data, which are difficult and expensive to obtain in routine practice. To reduce this dependence, we propose a stable semi-supervised virtual staining framework that jointly exploits both limited paired data and abundant unpaired source images. Directly incorporating unpaired images is challenging because their generated results lack corresponding targets for supervision, potentially leading to unrealistic staining, morphological degradation, or even training collapse. To obtain reliable supervision from these images, Hessian-derived morphology preservation extracts structural cues from each source image and constrains the generated output to retain tissue morphology. Histopathological realism constraints further guide the output toward plausible target-stain characteristics, preventing the source-derived structural supervision from degenerating into contour enhancement or simple color transformation. Together, the two components suppress structural and appearance drift, stabilize semi-supervised stain translation, and promote the preservation of diagnostically relevant information. Extensive experiments on H&E-to-IHC translation for Ki67 and HER2, as well as FFPE-to-H&E translation, demonstrate consistent improvements in image quality, morphology preservation, robustness, and downstream diagnostic performance. Code will be available.