T-MPEDNet: Unveiling the Synergy of Transformer-aware Multiscale Progressive Encoder-Decoder Network with Feature Recalibration for Tumor and Liver Segmentation

Chandravardhan Singh Raghaw, Jasmer Singh Sanjotra, Mohammad Zia Ur Rehman, Shubhi Bansal, Shahid Shafi Dar, Nagendra Kumar

Abstract: Highlights•A novel transformer-aware progressive encoder-decoder for segmenting liver and tumor.•Feature recalibration leverages channel-wise dependency to enhance spatial coherence.•Multi-scale feature extractor utilize large receptive field for fine-grained feature.•Efficiently pinpointing accurate boundaries with morphological erosion operation.•Quantitative and qualitative analyses demonstrate the superiority of T-MPEDNet.