MSAGHNet

This is the repository for 'DCCUNet: A Double Cross-Shaped Network for Pathology Image Segmentation'

Architecture

Dataset Structure

The dataset is organized as follows:

• data/

  • dataset_name/: Name of the dataset used, such as CoNIC, DSB, PanNuke, and MoNuSeg

    • train/: Contains training dataset
      • img/: Training images
      • mask/: Corresponding segmentation masks for training images
    • test/: Contains training dataset
      • img/: Test images
      • mask/: Corresponding segmentation masks for test images
    • val/: Contains validation dataset
      • img/: Validation images
      • mask/: Corresponding segmentation masks for validation images

  • dataset_name/: Name of the dataset used, such as CoNIC, DSB, PanNuke, and MoNuSeg

    • .......

Train and Test

Please use Train.py and Test.py for model training and prediction.

Datasets

The following datasets are used in this experiment:

1. CoNIC
2. DSB
3. PanNuke
4. MoNuSeg