Research map / Federated learning and privacy / Personalized and heterogeneous FL
Federated Split Task-Agnostic Vision Transformer for COVID-19 CXR Diagnosis
Sangjoon Park, Gwanghyun Kim, Jeongsol Kim, Boah Kim, Jong Chul Ye
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 multi-task learning (classification, segmentation, detection) to FESTA (ablation) | FESTA · COVID-19 CXR classification | average AUC: 0.909 -> 0.931 | Improved |
| add multi-task learning to FESTA (ablation) | FESTA · CXR segmentation | Dice: 0.798 -> 0.821 | Improved |
| add multi-task learning to FESTA (ablation) | FESTA · CXR detection | mAP: 0.202 -> 0.204 | No clear effect |
| add server-side Transformer body (ablation) | FESTA single-task · COVID-19 CXR classification | average AUC: 0.889 -> 0.909 | Improved |
| average heads/tails every 100 rounds instead of every 10 (ablation) | FESTA · COVID-19 CXR classification | average AUC: 0.822 -> 0.909 | Improved |
| average every 100 rounds instead of every 1000 (ablation) | FESTA · COVID-19 CXR classification | average AUC: 0.903 -> 0.909 | No clear effect |
| use FESTA (split ViT, head/tail federated) instead of plain federated learning | ViT classifier · COVID-19 CXR classification non-IID | average AUC: 0.891 -> 0.909 | Improved |
| use FESTA instead of split learning | ViT classifier · COVID-19 CXR classification non-IID | average AUC: 0.863 -> 0.909 | Improved |
Closest papers by meaning
- Chest ImaGenome Dataset for Clinical Reasoning (NeurIPS 2021 · similarity 0.82)
- FedVCK: Non-IID Robust and Communication-Efficient Federated Learning via Valuable Condensed Knowledge for Medical Image Analysis (AAAI 2025 · similarity 0.81)
- Point Transformer with Federated Learning for Predicting Breast Cancer HER2 Status from Hematoxylin and Eosin-Stained Whole Slide Images (AAAI 2024 · similarity 0.80)
- Addressing Asynchronicity in Clinical Multimodal Fusion via Individualized Chest X-ray Generation (NeurIPS 2024 · similarity 0.80)
- FedPIA – Permuting and Integrating Adapters Leveraging Wasserstein Barycenters for Finetuning Foundation Models in Multi-Modal Federated Learning (AAAI 2025 · similarity 0.80)
- Incongruent Multimodal Federated Learning for Medical Vision and Language-based Multi-label Disease Detection (AAAI 2025 · 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.
