AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Breaking Grid Constraints: Dynamic Graph Reconstruction Network for Multi-organ Segmentation

Junhao Xiao, Yang Wei, Jingyu Wang, Yongchao Wang, Xiuli Bi, Bin Xiao

ICCV 2025 · Graph learning › Graph self-supervised learning · approach: Brain and biological graph learning · 1 citations (OpenAlex)

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 changedWhereMeasuredEffect
replace grid-aggregation CNN/Transformer with dynamic graph reconstruction (DGRNet)DGRNet · BTCVDSC: 0.7777 (UTNet) -> 0.8657Improved
DGRNet versus best baselines on ACDC and CMRDGRNet · ACDC / CMRDSC: ACDC 0.9285 -> 0.9454; CMR 0.9032 -> 0.9171Improved
use EdgeConv instead of GIN/GraphSAGE/MR GConv (ablation)DGRNet · BTCVDSC: 0.8598 (MR GConv) -> 0.8657Improved
use DGR decoder instead of interpolation or deconvolution upsampling (ablation)DGRNet · BTCVDSC: 0.8353 (deconv) -> 0.8657Improved
use dynamic neighbor count 9 to 18 instead of fixed K (ablation)DGRNet · BTCVDSC: 0.8506 (K=21) -> 0.8657Improved
add category-perception loss Lcp with dynamic weighting (ablation)DGRNet · BTCVDSC: 0.8426 -> 0.8657Improved
dynamic instead of static loss ratio between Dice-CE and Lcp (ablation)DGRNet · BTCVDSC: 0.8582 -> 0.8657Improved
graph construction in DGRNetDGRNet · Model costFLOPs / Params: 10.8G FLOPs, 20.2M params versus TinyUNet 1.3G, 0.5MHurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy