Semi-Supervised Virtual Staining for High-Throughput and Label-Free Histopathology

Virtual staining techniques now enable the transformation of label-free images into clinically standard stained images; however, the extensive costs and time requirements of labelled dataset generation impede its widespread use in clinical settings. We propose a semi-supervised virtual staining for high-throughput and label-free histopathology that leverages both labeled and unlabeled datasets from stained and unstained tissues. We will further report the virtual staining results, compare the performance with conventional methods, and discuss future directions for our approach.

Duke Scholars

Author Roarke Horstmeyer Biomedical Engineering