AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

CCT-Net: Category-Invariant Cross-Domain Transfer for Medical Single-to-Multiple Disease Diagnosis

Yi Zhou, Lei Huang, Tao Zhou, Ling Shao

ICCV 2021 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Statistical alignment and hypothesis testing · 9 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
use confidence weighted pooling instead of average pooling (ablation)DSE-1 · ODIR-5K ocular classification (target, FGADR source)Cohen's kappa: 0.6556 -> 0.6712Improved
use CWP instead of LSE pooling (ablation)DSE-1 · ODIR-5K ocular classification (target, FGADR source)Cohen's kappa: 0.6561 -> 0.6712Improved
add CCT-Net refined features and DSE-2 (standard BN) (ablation)DSE-1-w-CWP · ODIR-5K ocular classification (target, FGADR source)Cohen's kappa: 0.6712 -> 0.7253Improved
use domain-specific batch normalization (ablation)DSE-1+CCT+DSE-2 · ODIR-5K ocular classification (target, FGADR source)Cohen's kappa: 0.7253 -> 0.7508Improved
add pixel-level supervision L_FR to the CCT pipeline (ablation)DSE-1+CCT+DSE-2 · ODIR-5K ocular classification (target, FGADR source)Cohen's kappa: 0.6747 -> 0.7253Improved
use CWP instead of average pooling for coarse maps (ablation)CCT-Net · FGADR lesion segmentation (MA, source domain)Dice: 0.493 -> 0.542Improved
use CWP instead of LSE pooling for coarse maps (ablation)CCT-Net · FGADR lesion segmentation (MA, source domain)Dice: 0.517 -> 0.542Improved
add CCT refinement to coarse CWP heatmaps (ablation)DSE-1 · FGADR lesion segmentation (MA, source domain)Dice: 0.453 -> 0.542Improved
use CCT-w-CWP instead of Attention U-NetAttention U-Net · FGADR lesion segmentation (MA, source domain)Dice: 0.536 -> 0.542Improved
use CWP instead of average pooling (ablation)DSE-1 · BrainTumor classification (meningioma)accuracy: 0.8863 -> 0.9067Improved
add CCT refinement and DSE-2 (standard BN) (ablation)DSE-1-w-CWP · BrainTumor classification (meningioma)accuracy: 0.9067 -> 0.9246Improved
use domain-specific BN (ablation)DSE-1+CCT+DSE-2 · BrainTumor classification (meningioma)accuracy: 0.9246 -> 0.9379Improved

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