TumorSegmentation

This repository provides the implementation of multiple segmentation models, including DAM-UNET, UNET, VNET, SegNet, SegResNet, and MaskRCNN, for the Medical Segmentation Decathlon (MSD) challenge focusing on Lung Tumorous Segmentation.

Steps for Using the Repository

1. Download the Repository:

   Clone or download the complete repository to your local machine.

2. Install Dependencies:

   Ensure all required dependencies are installed, including PyTorch, torchvision, seaborn, and others specified in the repository.

3. Download the Dataset:

   Obtain the MSD Lung Tumorous dataset (Task06_Lung.tar) from the following link: MSD Lung Tumorous Dataset.

   Extract the downloaded dataset and place it in the same directory as the repository code.

4. Preprocess the Dataset:

   Use the provided scripts Preprocessing.py and Preprocess-CT/Masks.py to preprocess the dataset. Adjust the paths for slices and masks in both scripts as needed to match your directory structure.

5. Train and Validate Models:

   Start training and validation for a specific model using the corresponding script. For example, to train the SegNet model, run the following command in your terminal:

   python SegNet.py

6. Check Results:

   Evaluate the results generated by the model and analyze its performance.

7. Repeat for Other Models:

   Repeat Steps 5 and 6 to train and visualize results for other models in the repository.