Research map / Detection and segmentation / CNN features
Class-Aware Adversarial Transformers for Medical Image Segmentation
Chenyu You, Ruihan Zhao, Fenglin Liu, Siyuan Dong, Sandeep P. Chinchali, ufuk topcu, Lawrence Hamilton Staib, James s Duncan
This paper tried a common change: Pretrained initialization
use pre-trained weights · Synapse multi-organ CT · DSC (%): 74.84 -> 82.17 Improved
What 86 accepted papers report for this change
- 74 improved
- 7 no clear effect
- 1 mixed
- 4 hurt
- Efficiently Training Time-to-First-Spike Spiking Neural Networks from Scratch ICML 2026 · initialize from pretrained ReLU weights · CIFAR10 · accuracy (%): 90.56 -> 93.89Improved
- Learning Visual Prior via Generative Pre-Training NeurIPS 2023 · initialize from pretrained GPT-2 text weights · COCO · KL divergence: 1.720 -> 1.959Hurt
- Mind-Omni: A Unified Multi-Task Framework for Brain-Vision-Language Modeling via Discrete Diffusion ICML 2026 · initialize from pre-trained Muddit instead of training from scratch · B→I&T description · BLEU1: 20.35 -> 29.12Improved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add class-aware transformer and transformer encoder modules to baseline (ablation) | TransUNet-like baseline · Synapse multi-organ CT | DSC (%): 77.48 -> 82.17 | Improved |
| add class-aware transformer module (vs w/o CAT) (ablation) | CATformer · Synapse multi-organ CT | DSC (%): 80.09 -> 82.17 | Improved |
| add transformer encoder module (vs w/o TEM) (ablation) | CATformer · Synapse multi-organ CT | DSC (%): 81.35 -> 82.17 | Improved |
| add WGAN-GP discriminator to CATformer (ablation) | CATformer · Synapse multi-organ CT | DSC (%): 82.17 -> 82.55 | Improved |
| add discriminator to CATformer (ablation) | CATformer · Synapse 95HD | 95HD: 16.20 -> 22.73 | Mixed |
| use pre-trained weights (ablation) | CATformer · Synapse multi-organ CT | DSC (%): 74.84 -> 82.17 | Improved |
| use pre-trained weights in both G and D (ablation) | CASTformer · Synapse multi-organ CT | DSC (%): 73.64 -> 82.55 | Improved |
| pre-train only G vs only D (ablation) | CASTformer · Synapse multi-organ CT | DSC (%): 78.87 -> 81.46 | Improved |
| use WGAN-GP loss instead of LS-GAN loss (ablation) | CASTformer · Synapse multi-organ CT | DSC (%): 81.45 -> 82.55 | Improved |
| use WGAN-GP loss instead of NS-GAN loss (ablation) | CASTformer · Synapse multi-organ CT | DSC (%): 80.02 -> 82.55 | Improved |
| increase iteration number N from 4 to 6 (ablation) | class-aware transformer module · Synapse multi-organ CT | DSC (%) | No clear effect |
| CASTformer vs TransUNet | segmentation model · LiTS | DSC (%): 67.94 -> 73.82 | Improved |
Closest papers by meaning
- ClassFormer: Exploring Class-Aware Dependency with Transformer for Medical Image Segmentation (AAAI 2023 · similarity 0.92)
- Combinatorial CNN-Transformer Learning with Manifold Constraints for Semi-supervised Medical Image Segmentation (AAAI 2024 · similarity 0.89)
- nnWNet: Rethinking the Use of Transformers in Biomedical Image Segmentation and Calling for a Unified Evaluation Benchmark (CVPR 2025 · similarity 0.88)
- DTMFormer: Dynamic Token Merging for Boosting Transformer-Based Medical Image Segmentation (AAAI 2024 · similarity 0.87)
- Generative Medical Segmentation (AAAI 2025 · similarity 0.86)
- DDViT: Double-Level Fusion Domain Adapter Vision Transformer (Student Abstract) (AAAI 2024 · similarity 0.86)
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.
