Research map / Federated learning and privacy / Personalized and heterogeneous FL
OmniFM: Toward Modality-Robust and Task-Agnostic Federated Learning for Heterogeneous Medical Imaging
Meilin Liu, Jiaying Wang, Jing Shan
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 |
|---|---|---|---|
| use OmniFM spectral retrieval framework instead of FedPer | FedPer · Cross-modality classification (Scenario 1, 20%) | accuracy (%): 84.47 -> 96.85 | Improved |
| use OmniFM instead of FedPer | FedPer · BreaKHis super-resolution (Scenario 1) | PSNR (dB): 32.99 -> 35.79 | Improved |
| use OmniFM instead of FedPer | FedPer · FeTS2022 segmentation | Dice (%): 73.86 -> 77.41 | Improved |
| use OmniFM instead of FedAvg | FedAvg · Multi-modal fusion (avg of 3 clients) | VIF: 0.279 -> 0.269 | Hurt |
| use OmniFM instead of FedPer | FedPer · Federated medical VQA Task 1 (SLAKE) | accuracy: F-2stage non-IID 0.824 -> 0.840; F-CL non-IID 0.833 -> 0.831 | Mixed |
| use OmniFM instead of FedPer | FedPer · Federated medical VQA Task 3 (Full Tuning) | average accuracy: 46.03 -> 47.30 | Improved |
| add Global Spectral Knowledge Retrieval (ablation) | OmniFM · Cross-modality classification (ablation) | accuracy (%): 95.82 -> 97.82 | Improved |
| add embedding-wise cross-attention (ablation) | OmniFM · Super-resolution (ablation) | PSNR (dB): 35.04 -> 35.79 | Improved |
| add prefix-suffix prompting (ablation) | OmniFM · Federated medical VQA Task 2 (ablation) | accuracy: 79.42 -> 79.27 | No clear effect |
| add prefix-suffix prompting (ablation) | OmniFM · Cross-modality classification (ablation) | accuracy (%): 96.92 -> 97.82 | Improved |
| add Spectral-Proximal Alignment (ablation) | OmniFM · Federated medical VQA Task 2 (ablation) | accuracy: 77.64 -> 79.27 | Improved |
| add Spectral-Proximal Alignment (ablation) | OmniFM · Federated medical VQA Task 3 (ablation) | accuracy: 42.37 -> 44.21 | Improved |
Closest papers by meaning
- Federated Modality-Specific Encoders and Multimodal Anchors for Personalized Brain Tumor Segmentation (AAAI 2024 · similarity 0.89)
- Incongruent Multimodal Federated Learning for Medical Vision and Language-based Multi-label Disease Detection (AAAI 2025 · similarity 0.87)
- FedVCK: Non-IID Robust and Communication-Efficient Federated Learning via Valuable Condensed Knowledge for Medical Image Analysis (AAAI 2025 · similarity 0.87)
- Overcoming Heterogeneous Data in Federated Medical Vision-Language Pre-training: A Triple-Embedding Model Selector Approach (AAAI 2025 · similarity 0.86)
- FedPIA – Permuting and Integrating Adapters Leveraging Wasserstein Barycenters for Finetuning Foundation Models in Multi-Modal Federated Learning (AAAI 2025 · similarity 0.85)
- Multi-modal Vision Pre-training for Medical Image Analysis (CVPR 2025 · similarity 0.85)
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.
