Research map / Vision-language and multimodal / Cross-modal learning
Virtual Nodes Guided Dynamic Graph Neural Network for Brain Tumor Segmentation with Missing Modalities
Sha Tao, Jiao Pan, Yu Guo, Chao Yao
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 |
|---|---|---|---|
| add virtual nodes (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | Dice enhancing tumor (%): 62.3 -> 66.0 | Improved |
| add virtual nodes (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | Dice tumor core (%): 79.4 -> 81.4 | Improved |
| add virtual nodes (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | Dice whole tumor (%): 86.2 -> 86.7 | Improved |
| add dynamic edge connection (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | Dice enhancing tumor (%): 64.6 -> 66.0 | Improved |
| add dynamic edge connection (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | Dice tumor core (%): 80.5 -> 81.4 | Improved |
| add dynamic edge connection (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | Dice whole tumor (%): 86.3 -> 86.7 | Improved |
| add heterogeneous weight matrix (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | Dice enhancing tumor (%): 62.9 -> 66.0 | Improved |
| add heterogeneous weight matrix (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | Dice tumor core (%): 80.9 -> 81.4 | Improved |
| increase virtual node length from 0 to 32 (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | average Dice (%): 75.9 -> 78.0 | Improved |
| increase virtual node length from 32 to 64 (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | average Dice (%): 78.0 -> 76.5 | Hurt |
| increase virtual node length from 32 to 128 (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | average Dice (%): 78.0 -> 76.0 | Hurt |
| increase virtual node length from 16 to 32 (ablation) | Virtual-node dynamic GNN · BraTS2018 (mean over 15 modality combos) | average Dice (%): 77.6 -> 78.0 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- M3AE — M3AE: Multimodal Representation Learning for Brain Tumor Segmentation with Missing Modalities (AAAI 2023) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Semantic-guided Masked Mutual Learning for Multi-modal Brain Tumor Segmentation with Arbitrary Missing Modalities (AAAI 2025 · similarity 0.89)
- Incomplete Multi-modal Brain Tumor Segmentation via Learnable Sorting State Space Model (CVPR 2025 · similarity 0.88)
- KMD: Koopman Multi-modality Decomposition for Generalized Brain Tumor Segmentation under Incomplete Modalities (CVPR 2025 · similarity 0.87)
- NaMa: Neighbor-Aware Multi-Modal Adaptive Learning for Prostate Tumor Segmentation on Anisotropic MR Images (AAAI 2024 · similarity 0.86)
- UniMRSeg: Unified Modality-Relax Segmentation via Hierarchical Self-Supervised Compensation (NeurIPS 2025 · similarity 0.86)
- Modality-agnostic Domain Generalizable Medical Image Segmentation by Multi-Frequency in Multi-Scale Attention (CVPR 2024 · similarity 0.85)
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.
