Data for CytoBulk

The uploaded package contains preprocessed data used in the CytoBulk paper for reproducibility. The reproduction directory is organized into several subfolders, including input/, result_data/, and model/ (or st_model/). Specifically, input/ stores the preprocessed datasets required to run the pipeline, model/ orst_model/ provides the trained model checkpoints saved during training, and result_data/ contains the final output results. In addition, DeepCMorph_Datasets_Combined_41_classes_acc_8159.pth is the pretrained model used in our experiments for H&E segmentation and cell type prediction.