MiniGlioma dataset for binary cell segmentation

This dataset consists solely of the Glioma C6-gen subset, originally designed to evaluate model generalization under varied imaging and seeding conditions. All images have been downsampled from the original resolution of 2592 × 1944 pixels to 704 × 512 pixels. It includes phase-contrast microscopy images of C6 glioma cells and corresponding binary segmentation masks in 8-bit .tif format. In these masks, pixels with a value of 255 represent cell regions, while pixels with a value of 0 correspond to the background. Unlike the full dataset, this version does not distinguish between cell types or nuclei—only general cell regions are annotated. The data is pre-split into two subsets: Training set: 25 images. Validation set: 5 images. Key characteristics: Data type: Phase-contrast microscopy images with binary masks. Microscopy: 2D phase-contrast images acquired at 24 or 72 hours after cell seeding. Microscope: BestScope BS-2092 in phase-contrast mode with 10× and 20× objectives. Cell type: C6 glioma cells (rat glial tumor cells, ATCC CCL-107). Image resolution: 704 × 512 pixels. File format: .tif (8-bit). File naming convention: Includes cultivation time and magnification, e.g., gen_24h_20x_17.tif Full dataset:The complete dataset, including both Glioma C6-spec and Glioma C6-gen subsets with detailed COCO-format annotations, is available at:https://zenodo.org/records/15083188 Authors: Roman Malashin¹ ², Svetlana Pashkevich³, Daniil Ilyukhin¹, Arseniy Volkov³, Valeria Yachnaya¹ ², Andrey Denisov³, Maria Mikhalkova¹ Affiliation(s): ¹ Pavlov Institute of Physiology, Russian Academy of Science ² Saint-Petersburg State University of Aerospace Instrumentation, Russia ³ Institute of Physiology, NAS of Belarus