MoRE-Net: An Interpretable and Modality-robust Model for Brain Tumor Grading

Key Points

• The aim is to create an interpretable model that accurately grades brain tumors despite missing modalities.
• Developed the MoRE-Net model for brain tumor grading.
• Tested the model under conditions where certain data modalities were missing.
• Evaluated both diagnostic accuracy and interpretability of the model.
• MoRE-Net achieved high diagnostic accuracy in brain tumor grading.
• The model maintained interpretability even when faced with missing modalities.
• Promising results suggest feasibility for clinical applications.