Personalized Longitudinal Medical Report Generation via Temporally-Aware Federated Adaptation
This work addresses the problem of modeling temporal shifts and patient heterogeneity in medical report generation for healthcare applications, offering a privacy-preserving solution, though it is incremental as it builds on existing federated learning methods.
The paper tackled the challenge of generating personalized longitudinal medical reports under privacy constraints by proposing FedTAR, a federated learning framework that integrates demographic-driven personalization with time-aware aggregation, resulting in consistent improvements in linguistic accuracy, temporal coherence, and cross-site generalization on datasets like J-MID (1M exams) and MIMIC-CXR.
Longitudinal medical report generation is clinically important yet remains challenging due to strict privacy constraints and the evolving nature of disease progression. Although federated learning (FL) enables collaborative training without data sharing, existing FL methods largely overlook longitudinal dynamics by assuming stationary client distributions, making them unable to model temporal shifts across visits or patient-specific heterogeneity-ultimately leading to unstable optimization and suboptimal report generation. We introduce Federated Temporal Adaptation (FTA), a federated setting that explicitly accounts for the temporal evolution of client data. Building upon this setting, we propose FedTAR, a framework that integrates demographic-driven personalization with time-aware global aggregation. FedTAR generates lightweight LoRA adapters from demographic embeddings and performs temporal residual aggregation, where updates from different visits are weighted by a meta-learned temporal policy optimized via first-order MAML. Experiments on J-MID (1M exams) and MIMIC-CXR demonstrate consistent improvements in linguistic accuracy, temporal coherence, and cross-site generalization, establishing FedTAR as a robust and privacy-preserving paradigm for federated longitudinal modeling.