Training- and Test-Dataset

The dataset provides paired image data designed for training and testing of deep learning models for virtual staining in fluorescence microscopy. It includes grayscaled input images and their corresponding ground truth fluorescence images stained with DAPI (nuclei) and Phalloidin (cytoskeleton) of osteoclast cultures, enabling supervised training.The images were acquired using a Stellaris 8 microscope (Leica Microsystems GmbH) under consistens imaging conditions. The training set contains 4.515 image pairs (grayscale input image + corresponding RGB fluorescence image). The test set includes additional 210 image pairs.All images are provided in uncompressed TIFF format with a fixed size of 256x256 pixels.