Research map / Vision-language and multimodal / Cross-modal learning
Foundation-Guided Representation Alignment for Multimodal Medical Image Registration
Mengjie Guo, Xinxing Cheng, Wenqi Lu, Qingjie Meng, Guanyu Yang, Yang Chen, Ziyun Ding, Alejandro Frangi, Jinming Duan
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 representation alignment with a frozen DINOv3 to the ModeT encoder (repeated-slice input) (ablation) | ModeT · SR-Reg brain MR-CT | DSC (%): 72.30 -> 76.83 | Improved |
| add representation alignment (ablation) | ModeT · SR-Reg brain MR-CT | HD95: 2.58 -> 2.23 | Improved |
| add Data ReAssembly instead of repeating a slice three times (ablation) | ModeT+RA · SR-Reg brain MR-CT | DSC (%): 76.83 -> 77.26 | Improved |
| add Data ReAssembly (ablation) | ModeT+RA · SR-Reg brain MR-CT | HD95: 2.23 -> 2.16 | Improved |
| add Hierarchical Representation Fusion instead of final-layer features (ablation) | ModeT+RA · SR-Reg brain MR-CT | DSC (%): 76.83 -> 78.08 | Improved |
| add Hierarchical Representation Fusion (ablation) | ModeT+RA · SR-Reg brain MR-CT | HD95: 2.23 -> 2.09 | Improved |
| combine DR and HRF with RA (full model) (ablation) | ModeT+RA · SR-Reg brain MR-CT | DSC (%): 78.08 -> 79.04 | Improved |
| add Data ReAssembly and HRF to representation alignment (ablation) | ModeT+RA · SR-Reg brain MR-CT | %(|Jphi|<=0): 0.01 -> 0.06 | Hurt |
| add foundation-guided alignment to ModeT | ModeT · Learn2Reg abdomen MR-CT | DSC (%): 73.84 -> 76.16 | Improved |
| add foundation-guided alignment to ModeT | ModeT · Learn2Reg abdomen MR-CT | HD95 (mm): 7.80 -> 7.12 | Improved |
| add foundation-guided alignment to PiViT | PiViT · Learn2Reg abdomen MR-CT | DSC (%): 67.15 -> 71.26 | Improved |
| add foundation-guided alignment to PiViT | PiViT · Learn2Reg abdomen MR-CT | HD95 (mm): 11.55 -> 10.47 | Improved |
Closest papers by meaning
- Modality-Agnostic Structural Image Representation Learning for Deformable Multi-Modality Medical Image Registration (CVPR 2024 · similarity 0.88)
- Unsupervised Multi-modal Medical Image Registration via Invertible Translation (ECCV 2024 · similarity 0.88)
- Towards Cross-Modality Medical Image Segmentation with Online Mutual Knowledge Distillation (AAAI 2020 · similarity 0.88)
- Multi-modal Vision Pre-training for Medical Image Analysis (CVPR 2025 · similarity 0.87)
- When to Align, When to Predict: A Phase Diagram for Multimodal Learning (NeurIPS 2026 · similarity 0.85)
- Unlocking the Power of Spatial and Temporal Information in Medical Multimodal Pre-training (ICML 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.
