OmniMed-FL: A Robust Multimodal Federated Learning Framework for Clinical Diagnosis
OmniMed-FL benchmarks multimodal federated learning for chest radiograph diagnosis across 3-20 clients, with FedProx leading under severe non-IID skew.
OmniMed-FL studies multimodal federated learning combining chest radiographs and clinical notes for five-class condition classification under HIPAA/GDDR-compliant decentralized training. It benchmarks eight fusion strategies, imputation rules, and federated baselines under Dirichlet non-IID partitioning across 3-20 hospital clients. With 5 clients and severe skew (alpha=0.1), FedProx scored 0.737 macro-F1 versus 0.662 for FedAvg and 0.297 for local-only training. Multimodal fusion beat unimodal inputs (0.956 vs 0.934 text, 0.664 images) on the synthetic corpus.
- Controlled study of image+text federated learning across 3-20 clients
- FedProx (0.737 F1) outperforms FedAvg (0.662) and local-only (0.297) under skew
- Label skew costs up to 0.27 F1, more than sevenfold client increase (0.10)
- Descriptive proxy comparisons; notes are synthetic, not patient-level
From the original article
1 min read at arxiv.orgSimultaneous assessment of medical imaging and patient records is often required in clinical diagnosis. However, standard machine learning algorithms cannot analyze these data types together. Meanwhile, compliance with HIPAA and GDPR can constrain centralized aggregation of sensitive patient data.
By Ayush Debnath · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.