AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Anatomy-aware Representation Learning for Medical Ultrasound

Seok-Hwan Oh, Myeong-Gee Kim, Guil Jung, hyeonjik lee, Young-Min Kim, Sang-yun Kim, Hyuksool Kwon, Hyeonmin Bae

ICLR 2026 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning

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
pretrain MIM on ultrasound images instead of natural images (ablation)ViT-B MIM · BUSI breast canceraccuracy (%): 83.09 -> 89.43Improved
add self-distillation loss to MIM on ultrasound (ablation)A-ViT w/o ACDT · BUSI breast canceraccuracy (%): 89.43 -> 92.25Improved
add adversarial loss for speckle preservation (ablation)A-ViT · BUSI breast canceraccuracy (%): 92.25 -> 92.95Improved
add anatomy-conditioned deformable transformer (ACDT) (ablation)A-ViT · BUSI breast canceraccuracy (%): 92.95 -> 93.66Improved
use ultrasound-specific ARL pretraining instead of natural-image SSL (MAE)ViT-B · BUSI breast cancer (linear probing)accuracy (%): 77.64 -> 86.62Improved
reduce cardiac view training data to 1% (ablation)A-ViT vs baselines · Cardiac view classificationTop-1 accuracyNo clear effect
use A-ViT versus best SSL baseline on COVID-19 lung USA-ViT · COVID-19 classificationaccuracy (%): 87.67 -> 91.44Improved

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