AI Paper Atlas

Research map / Detection and segmentation / CNN features

Keypoint-Augmented Self-Supervised Learning for Medical Image Segmentation with Limited Annotation

Zhangsihao Yang, Mengwei Ren, Kaize Ding, Guido Gerig, Yalin Wang

NeurIPS 2023 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 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 KAF layers at all four encoder scales (ablation)UNet (random init) · CHD M=15Dice: 0.627 -> 0.686Improved
add KAF layers at all four encoder scales (ablation)UNet (random init) · ACDC M=6Dice: 0.782 -> 0.827Improved
remove KAF at last scale l4 (Exp B vs F) (ablation)KAF-UNet · CHD M=15Dice: 0.686 -> 0.658Hurt
reduce self-attention layers in KAF transformer from 9 to 3 (ablation)KAF-UNet · CHD M=15Dice: 0.686 -> 0.677Hurt
reduce self-attention layers from 9 to 6 (ablation)KAF-UNet · CHD M=15Dice: 0.686 -> 0.690No clear effect
combine global conv and keypoint SSL losses vs single global loss (ablation)KAF-UNet SSL · CHD M=15Dice: 0.699 -> 0.711Improved
add local correspondence loss w3=0.01 (ablation)KAF-UNet SSL · CHD M=15Dice: 0.711 -> 0.712Improved
add local loss w3=0.01 on ACDC (ablation)KAF-UNet SSL · ACDC M=6Dice: 0.865 -> 0.873Improved
replace SIFT with SuperPoint keypoints (random init) (ablation)KAF-UNet · CHD M=2Dice: 0.686 -> 0.643Hurt
replace SIFT with SuperPoint keypoints (SSL) (ablation)KAF-UNet SSL · CHD M=2Dice: 0.712 -> 0.703No clear effect
vary keypoint correspondence threshold 5 to 40 (ablation)KAF-UNet SSL local loss · CHD M=15Dice: 0.684 -> 0.690No clear effect
widen UNet with duplicated channels (UNet(c2)) vs standard UNet (ablation)UNet · CHD M=15Dice: 0.627 -> 0.646Improved

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