Research map / Detection and segmentation / CNN features
Fractal-G: Topology-Aware Heterogeneous Graphs for Medical Image Segmentation
Bai Sun, Zixuan Tang
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes Fractal-G, a plug-and-play multi-scale fusion module using a Grid-Graph-Grid feature pipeline to combine macroscopic semantics with fine boundary geometry for medical image segmentation.
Problem
Medical image segmentation must align low-frequency macroscopic semantic context with sparse, high-frequency boundary details, and existing representations struggle to balance these.
Approach
A continuous Grid-Graph-Grid feature pipeline with topology-aware heterogeneous graphs and a fractal-style multi-scale fusion, designed as a drop-in module to bridge the information asymmetry between semantics and boundaries.
Closest papers by meaning
- Topology-Reinforced Swin Transformer for Medical Image Analysis (NeurIPS 2026 · similarity 0.86)
- Topograph: An Efficient Graph-Based Framework for Strictly Topology Preserving Image Segmentation (ICLR 2025 · similarity 0.86)
- Neighbor Does Matter: Density-Aware Contrastive Learning for Medical Semi-supervised Segmentation (AAAI 2025 · similarity 0.85)
- S2S2: Semantic Stacking for Robust Semantic Segmentation in Medical Imaging (AAAI 2025 · similarity 0.84)
- MedFlowSeg: Flow Matching for Medical Image Segmentation with Frequency-Aware Attention (NeurIPS 2026 · similarity 0.84)
- DyCON: Dynamic Uncertainty-aware Consistency and Contrastive Learning for Semi-supervised Medical Image Segmentation (CVPR 2025 · 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.
