Research map / Detection and segmentation / CNN features
Segmenting Medical MRI via Recurrent Decoding Cell
Ying Wen, Kai Xie, Lianghua He
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace U-Net decoding with recurrent decoding | U-Net-like backbone · BrainWeb | Dice: 0.9892 -> 0.9934 | Improved |
| replace U-Net decoding with recurrent decoding | VGG16 · MRBrainS | Dice: 0.8991 -> 0.9126 | Improved |
| replace U-Net decoding with recurrent decoding | VGG16 · HVSMR | Dice: 0.8201 -> 0.8813 | Improved |
| use a shared-parameter recurrent decoder | VGG16 encoder-decoder · 240×240×3 input | number of parameters: 25.86M -> 14.87M | Improved |
| use a shared-parameter recurrent decoder | U-Net-like encoder-decoder · 240×240×3 input | number of parameters: 1.94M -> 1.23M | Improved |
| increase noise from 1% to 9% | CRDN with U-Net-like backbone · BrainWeb | Dice coefficient reduction: 1.69% | Hurt |
| increase intensity non-uniformity from 0% to 40% | CRDN with U-Net-like backbone · BrainWeb | Dice coefficient reduction: 0.23% | Hurt |
| use ConvLSTM instead of ConvRNN (ablation) | RDC with VGG16 · MRBrainS | Dice: 0.9088 -> 0.9126 | Improved |
| use ConvLSTM instead of ConvRNN (ablation) | RDC with VGG16 · HVSMR | Dice: 0.8813 -> 0.8641 | Hurt |
| use ConvRNN instead of ConvLSTM (ablation) | RDC with U-Net-like backbone · BrainWeb | Dice: 0.9919 -> 0.9934 | Improved |
| use ConvLSTM instead of ConvRNN (ablation) | RDC with U-Net-like backbone · MRBrainS | Dice: 0.9068 -> 0.9112 | Improved |
| use ConvRNN instead of ConvGRU (ablation) | RDC with ResNet50 · HVSMR | Dice: 0.8696 -> 0.8641 | Hurt |
Closest papers by meaning
- nnWNet: Rethinking the Use of Transformers in Biomedical Image Segmentation and Calling for a Unified Evaluation Benchmark (CVPR 2025 · similarity 0.85)
- Revisiting MAE Pre-training for 3D Medical Image Segmentation (CVPR 2025 · similarity 0.84)
- EffiDec3D: An Optimized Decoder for High-Performance and Efficient 3D Medical Image Segmentation (CVPR 2025 · similarity 0.84)
- Learning Generalized Medical Image Segmentation from Decoupled Feature Queries (AAAI 2024 · similarity 0.84)
- E2ENet: Dynamic Sparse Feature Fusion for Accurate and Efficient 3D Medical Image Segmentation (NeurIPS 2024 · similarity 0.84)
- Unleashing the Potential of SAM for Medical Adaptation via Hierarchical Decoding (CVPR 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
