Research map / Architectures and efficiency / Attention and transformers
RadioTransformer: A Cascaded Global-Focal Transformer for Visual Attention-Guided Disease Classification
Moinak Bhattacharya, Shubham Jain, Prateek Prasanna
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add visual-attention guidance (HVAT+VAL) (ablation) | RadioTransformer · RSNA-MIDRC | AUC: 71.78 -> 79.60 | Improved |
| add visual-attention guidance (HVAT+VAL) (ablation) | RadioTransformer · RSNA-MIDRC | accuracy: 50.17 -> 57.07 | Improved |
| add visual-attention guidance (HVAT+VAL) (ablation) | RadioTransformer · COVID-19 Radiography | accuracy: 98.51 -> 99.43 | Improved |
| add visual-attention guidance (HVAT+VAL) (ablation) | RadioTransformer · RSNA Pneumonia | accuracy: 98.19 -> 98.94 | Improved |
| add visual-attention guidance (HVAT+VAL) (ablation) | RadioTransformer · SIIM COVID-19 | accuracy: 97.47 -> 98.10 | Improved |
| add visual-attention guidance (HVAT+VAL) (ablation) | RadioTransformer · Cell Pneumonia | AUC: 89.82 -> 88.80 | Hurt |
| add visual-attention guidance (HVAT+VAL) (ablation) | RadioTransformer · Cell Pneumonia | accuracy: 82.05 -> 80.73 | Hurt |
| add visual-attention guidance (HVAT+VAL) (ablation) | RadioTransformer · NIH Chest X-rays | AUC: 85.48 -> 85.43 | No clear effect |
| add visual-attention guidance (HVAT+VAL) (ablation) | RadioTransformer · TCIA-SBU | AUC: 93.75 -> 94.76 | Improved |
| use RadioTransformer instead of Swin1 | Swin1 · TCIA-SBU | AUC: 73.68 -> 94.76 | Improved |
| use RadioTransformer instead of Swin1 | Swin1 · NIH Chest X-rays | AUC: 74.18 -> 85.43 | Improved |
Closest papers by meaning
- A World Model of Radiologist Reading for Medical Image Representation Learning (NeurIPS 2026 · similarity 0.80)
- CARE Transformer: Mobile-Friendly Linear Visual Transformer via Decoupled Dual Interaction (CVPR 2025 · similarity 0.80)
- Statistical Test for Attention Maps in Vision Transformers (ICML 2024 · similarity 0.80)
- G2D: From Global to Dense Radiography Representation Learning via Vision-Language Pre-training (NeurIPS 2024 · similarity 0.80)
- Less Attention is More: Prompt Transformer for Generalized Category Discovery (CVPR 2025 · similarity 0.80)
- You Only Need Less Attention at Each Stage in Vision Transformers (CVPR 2024 · similarity 0.80)
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.
