AI Paper Atlas

Research map / Detection and segmentation / CNN features

Difficulty-Aware Attention Network with Confidence Learning for Medical Image Segmentation

Dong Nie, Li Wang, Lei Xiang, Sihang Zhou, Ehsan Adeli, Dinggang Shen

AAAI 2019 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 45 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 difficulty-aware attention mechanism (ablation)enUNet · Pelvic prostate baseDSC (%): 86.70 -> 89.41Improved
add difficulty-aware attention mechanism (ablation)enUNet · Pelvic prostate middleDSC (%): 87.91 -> 90.90Improved
add difficulty-aware attention mechanism (ablation)enUNet · Pelvic prostate apexDSC (%): 83.92 -> 88.21Improved
use focal loss (ablation)enUNet · Pelvic prostate apexDSC (%): 83.92 -> 86.83Improved
use focal loss (ablation)enUNet · Pelvic prostate baseDSC (%): 86.70 -> 88.24Improved
replace focal loss with difficulty-aware attention (ablation)enUNet · Pelvic prostate apexDSC (%): 86.83 -> 88.21Improved
replace hybrid loss with difficulty-aware attention (ablation)enUNet · Pelvic prostate middleDSC (%): 90.08 -> 90.90Improved
use hybrid Dice+CE loss instead of focal loss (ablation)enUNet · Pelvic prostate apexDSC (%): 86.83 -> 86.71No clear effect
add enhancements (residual, deep supervision, attention fusion) to UNet (ablation)UNet · Pelvic MRIDSC (%)Improved
add difficulty-aware attention mechanism (ablation)enUNet · MR Brain challengeDSC (%): about 3.5% average DSCImproved
proposed single model vs top ensemble methodsProposed · PROMISE12 prostate challengeoverall score: 88.28(3.02) vs 88.98(3.41)Hurt

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