Pixel super-resolved virtual staining of label-free tissue using diffusion models

Virtual staining of tissue offers a powerful tool for transforming label-free microscopy images of unstained tissue into equivalents of histochemically stained samples. This study presents a diffusion model-based pixel super-resolution virtual staining approach utilizing a Brownian bridge process to enhance both the spatial resolution and fidelity of label-free virtual tissue staining, addressing the limitations of traditional deep learning-based methods. Our approach integrates sampling techniques into a diffusion model-based image inference process to significantly reduce the variance in the generated virtually stained images, resulting in more stable and accurate outputs. Blindly applied to lower-resolution auto-fluorescence images of label-free human lung tissue samples, the diffusion-based pixel super-resolution virtual staining model consistently outperforms conventional approaches in resolution, structural similarity and perceptual accuracy, successfully achieving a pixel super-resolution factor of 4-5×, increasing the output space-bandwidth product by 16-25-fold compared to the input label-free microscopy images. Diffusion-based pixel super-resolved virtual tissue staining not only improves resolution and image quality but also enhances the reliability of virtual staining without traditional chemical staining, offering significant potential for clinical diagnostics. Virtual staining transforms label-free microscopy images of unstained tissue into their histochemically stained equivalents. Here, authors introduce a diffusion model-based pixel super-resolution virtual staining method to improve the spatial resolution and accuracy of label-free virtual staining.