MoME: A Foundation Model for Brain Lesion Segmentation with Mixture of Modality Experts
This model is designed for brain lesion segmentation using the MoE architecture, modified within the nnU-Net framework.
For detailed usage and instructions, please refer to the GitHub repository.
We have released both the pretrained experts (for T1, T1ce, T2, FLAIR, and DWI) and the final well-trained MoME model.
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no library tag.