Lung Cancer Histopathology Images (1000x1000 pixel crops)

This dataset contains cropped single-cell image patches extracted from hematoxylin and eosin (H&E)-stained histopathology sections of human lung squamous cell carcinoma. The original material was obtained from digitized whole-slide images of formalin-fixed paraffin-embedded tissue sections. Each image is a 1000 × 1000 pixel cell-centered crop. The phenotype label assigned to each crop refers specifically to the cell located at the center of the image, while the surrounding tissue context is preserved to capture the local microenvironment. The dataset is organized into phenotype-specific directories, where the folder names correspond to the assigned class labels. The annotated phenotypic categories are: Chondrocytes Lymphocytes Mitotic_SCC_cell Necrosis Other_non_neopl_cells Plasma_cells SCC_cell These data were generated for the study of microenvironment-aware single-cell phenotype classification in lung cancer histopathology and are associated with the manuscript: “Microenvironmental information significantly improves the recognition of cell types in human lung cancer patients.”