Research map / Label-efficient and robust learning / Self-supervised and contrastive
VasoMIM: Vascular Anatomy-Aware Masked Image Modeling for Vessel Segmentation
De-Xing Huang, Xiao-Hu Zhou, Mei-Jiang Gui, Xiao-Liang Xie, Shi-Qi Liu, Shuang-Yi Wang, Tian-Yu Xiang, Rui-Ze Ma, Nu-Fang Xiao, Zeng-Guang Hou
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 anatomy-guided masking (ablation) | MAE (gamma 0.5) · ARCADE | DSC (%): 68.00 -> 68.30 | Improved |
| add anatomy-guided masking (ablation) | MAE (gamma 0.5) · CAXF | DSC (%): 83.15 -> 83.96 | Improved |
| add anatomical consistency loss (ablation) | MAE (gamma 0.5) · ARCADE | DSC (%): 68.00 -> 68.45 | Improved |
| add anatomical consistency loss (ablation) | MAE (gamma 0.5) · CAXF | DSC (%): 83.15 -> 84.03 | Improved |
| combine anatomy-guided masking and consistency loss (ablation) | MAE (gamma 0.5) · CAXF | DSC (%): 83.15 -> 84.49 | Improved |
| add anatomical consistency loss (ablation) | VasoMIM representations · XCAD clustering | Calinski-Harabasz Index: 17.11 -> 607.24 | Improved |
| use weak-to-strong guidance (beta 0 to 0.5) instead of random masking (ablation) | VasoMIM · ARCADE | DSC (%): 68.45 -> 68.85 | Improved |
| use constant strongest anatomical guidance (beta 1 to 1) (ablation) | VasoMIM · ARCADE | DSC (%): 68.45 -> 65.36 | Hurt |
| reverse schedule to strong-to-weak guidance (ablation) | VasoMIM · ARCADE | DSC (%): 68.45 -> 67.81 | Hurt |
| use masking ratio 0.5 instead of 0.75 (ablation) | VasoMIM · ARCADE | DSC (%) | Improved |
| MIM pre-training with VasoMIM instead of supervised segmentor pre-training on Frangi pseudo-labels (ablation) | U-Net · ARCADE | DSC (%): 66.74 -> 68.85 | Improved |
| MIM pre-training with VasoMIM instead of supervised segmentor pre-training on Frangi pseudo-labels (ablation) | U-Net · CAXF | DSC (%): 83.18 -> 84.49 | Improved |
Closest papers by meaning
- vesselFM: A Foundation Model for Universal 3D Blood Vessel Segmentation (CVPR 2025 · similarity 0.87)
- DeNVeR: Deformable Neural Vessel Representations for Unsupervised Video Vessel Segmentation (CVPR 2025 · similarity 0.85)
- VasTSD: Learning 3D Vascular Tree-state Space Diffusion Model for Angiography Synthesis (CVPR 2025 · similarity 0.85)
- Why Invariance is Not Enough for Biomedical Domain Generalization and How to Fix It (NeurIPS 2026 · similarity 0.82)
- Representing Part-Whole Hierarchies in Foundation Models by Learning Localizability Composability and Decomposability from Anatomy via Self Supervision (CVPR 2024 · similarity 0.81)
- Augmented Equivariant Mesh Networks for Anatomical Segmentation (NeurIPS 2026 · similarity 0.81)
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.
