In this research work, to address missing modality issue in the MRI-based semantic segmentation tasks, we propose MMCFormer, a novel missing modality compensation network. Our strategy builds upon 3D efficient transformer blocks and uses a co-training strategy to effectively train a missing modality network. To ensure feature consistency in a multi-scale fashion, MMCFormer utilizes global contextual agreement modules in each scale of the encoders. Furthermore, to transfer modality-specific representations, we propose to incorporate auxiliary tokens in the bottleneck stage to model interaction between full and missing-modality paths. On top of that, we include feature consistency losses to reduce the domain gap in network prediction and increase the prediction reliability for the missing modality path.

• Paper accepted in MIDL2023

This code has been implemented in Python language using Pytorch library and tested in Ubuntu OS, though should be compatible with related environment. following environment and Library needed to run the code:

• Python 3
• Pytorch

For training deep model and evaluating the BraTA 2018 dataset set, follow the bellow steps:
1- Download the BraTS 2018 train dataset from this link and extract it inside the dataset_BraTS2018 folder.
2- Run train.py for training the model.
3- For performance calculation and producing segmentation results, run evaluation.ipynb.

Notice: our implementation uses the SMU codes: https://github.com/rezazad68/smunet

If this code helps with your research please consider citing the following paper:

@inproceedings{
  karimijafarbigloo2023mmcformer,
  title={{MMCF}ormer: Missing Modality Compensation Transformer for Brain Tumor Segmentation},
  author={Sanaz Karimijafarbigloo and Reza Azad and Amirhossein Kazerouni and Saeed Ebadollahi and Dorit Merhof},
  booktitle={Medical Imaging with Deep Learning},
  year={2023},
  url={https://openreview.net/forum?id=PD0ASSmvlE}
}