ReFRM3D: A Radiomics-enhanced Fused Residual Multiparametric 3D Network with Multi-Scale Feature Fusion for Glioma Characterization
ReFRM3D is a new 3D network for characterizing gliomas using radiomics. It integrates multi-scale feature fusion, hybrid upsampling, and an extended residual skip mechanism within a 3D U-Net architecture. This approach improves brain tumor segmentation accuracy across BraTS datasets while providing detailed morphological insights through extracted radiomic features.