Research map / Label-efficient and robust learning / Semi-supervised learning
RA-BUSSeg: Relation-aware Semi-supervised Breast Ultrasound Image Segmentation via Adjacent Propagation and Cross-layer Alignment
Wanting Zhang, Zhenhui Ding, Guilian Chen, Huisi Wu, Jing Qin
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 adjacent relation propagation (ARP) (ablation) | SupOnly baseline · UDIAT 1/4 labeled | Dice: 75.11 -> 80.79 | Improved |
| add adjacent relation propagation (ARP) (ablation) | SupOnly baseline · BUSI 1/4 labeled | Dice: 71.62 -> 76.31 | Improved |
| add cross-layer relation alignment (CRA) (ablation) | SupOnly baseline · UDIAT 1/4 labeled | Dice: 75.11 -> 81.21 | Improved |
| add cross-layer relation alignment (CRA) (ablation) | SupOnly baseline · BUSI 1/4 labeled | Dice: 71.62 -> 76.42 | Improved |
| add CRA on top of ARP (ablation) | RA-BUSSeg · UDIAT 1/4 labeled | Dice: 80.79 -> 82.55 | Improved |
| add CRA on top of ARP (ablation) | RA-BUSSeg · BUSI 1/4 labeled | Dice: 76.31 -> 77.95 | Improved |
| add ARP on top of CRA (ablation) | RA-BUSSeg · UDIAT 1/4 labeled | IoU: 72.42 -> 74.21 | Improved |
| add ARP on top of CRA (ablation) | RA-BUSSeg · BUSI 1/4 labeled | IoU: 67.84 -> 69.30 | Improved |
| use RA-BUSSeg instead of strongest BUS-specific baseline PH-Net | semi-supervised BUS segmentation · BUSI 1/4 labeled | Dice: 76.81 -> 77.95 | Improved |
| use RA-BUSSeg instead of ABD | semi-supervised segmentation · BUSI | Dice: outperforms ABD by 1.46%, 1.65%, and 1.35% in Dice | Improved |
| use a 3x3 sliding window for ARP rather than larger (ablation) | ARP · UDIAT and BUSI 1/4 labeled | Dice: optimal performance with a 3 × 3 sliding window | Improved |
| use global propagation instead of adjacent propagation (ablation) | ARP · BUSI/UDIAT | boundary quality | Hurt |
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.
- PH-Net — PH-Net: Semi-Supervised Breast Lesion Segmentation via Patch-wise Hardness (CVPR 2024) · better on 8 of 8 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
- CSC-PA: Cross-image Semantic Correlation via Prototype Attentions for Single-network Semi-supervised Breast Tumor Segmentation (CVPR 2025 · similarity 0.90)
- PH-Net: Semi-Supervised Breast Lesion Segmentation via Patch-wise Hardness (CVPR 2024 · similarity 0.88)
- Semi-supervised TEE Segmentation via Interacting with SAM Equipped with Noise-Resilient Prompting (AAAI 2024 · similarity 0.86)
- Anatomical Knowledge Mining and Matching for Semi-supervised Medical Multi-structure Detection (AAAI 2025 · similarity 0.85)
- MAPSeg: Unified Unsupervised Domain Adaptation for Heterogeneous Medical Image Segmentation Based on 3D Masked Autoencoding and Pseudo-Labeling (CVPR 2024 · similarity 0.85)
- Weakly Supervised Gland Segmentation with Class Semantic Consistency and Purified Labels Filtration (AAAI 2025 · 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.
