Automatic Liver Tumor Segmentation on Dynamic Contrast Enhanced MRI Using 4D Information: Deep Learning Model Based on 3D Convolution and Convolutional LSTM

This is the source code for liver tumor segmentation with 3D CNN and Conv-LSTM model.

1. Data pre-processing for liver segmentation:

   Run python3 Data_prepropcess.py

2. Liver segmentation with 3D U-net model:

   Run python3 Liver_seg_model.py

3. Liver patch extraction:

   Run python3 Liver_patch_extraction.py

4. Tumor patch extraction:

   Run python3 Tumor_patch_extraction.py

5. Tumor segmentation with combined CNN and Conv-LSTM model:

   2.5D version (2D CNN + Conv-LSTM): Run python3 Tumor_seg_model_2DCT.py

   3D version (3D CNN + Conv-LSTM): Run python3 Tumor_seg_model_3DCT.py