AI Paper Atlas

Research map / Federated learning and privacy / Personalized and heterogeneous FL

DSFedMed: Dual-Scale Federated Medical Image Segmentation via Mutual Distillation Between Foundation and Lightweight Models

Hanwen Zhang, Qiaojin Shen, Yuxi Liu, Yuesheng Zhu, Guibo Luo

AAAI 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated medical and multimodal 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
add mutual KD (ablation)DSFedMed · ProstateDice: 0.780 -> 0.791Improved
add learnability-guided selection (ablation)DSFedMed · ProstateDice: 0.780 -> 0.794Improved
add learnability-guided selection (ablation)DSFedMed · training timetime (s): 399.010 -> 209.827Improved
add both mutual KD and LG selection (ablation)DSFedMed · ProstateDice: 0.780 -> 0.813Improved
add both mutual KD and LG selection (ablation)DSFedMed · CHAOSDice: 0.893 -> 0.927Improved
add mutual KD alone vs LG selection alone (ablation)DSFedMed · CHAOSDice: 0.916 (KD) vs 0.911 (LG) vs 0.893 noneMixed
use TinySAM client instead of U-Net (ablation)DSFedMed · Fundus-OCDice: 0.739 -> 0.797Improved
replace SAM-B server with larger MSA-L (ablation)DSFedMed · Fundus-OCDice: 0.797 -> 0.795No clear effect
use ControlNet instead of DDPM or Pix2Pix generator (ablation)DSFedMed · Fundus generated dataFID: 34.86 -> 7.83Improved
DSFedMed vs FedSAMDSFedMed · Average over 5 datasetsDice and communication: 0.810 -> 0.829; 71,538 MB -> 8,920 MBImproved
DSFedMed vs FedSAM on ProstateDSFedMed · ProstateDice: 0.814 -> 0.813No clear effect
select samples with rate 0.2-0.8 and lambda 0.1-0.9 (ablation)DSFedMed · five datasetsDiceImproved

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