Get Started
Home
Topics
Search
Library
Research questionHow can medical concept encoders adapt text-attributed knowledge-graph context to each patient for EHR prediction?A clinical code can have different meaning and predictive value across patients and their trajectories. Models must determine how much relational context each code needs while connecting that structure with its textual semantics.
AI
Health
Machine Learning
Natural Language Processing
Reinforcement Learning
Research Paper
Latest papersRecent research connected to this question, newest first.REFINE: LLM Refinement over Budgeted Text-Attributed Graphs for Personalized Medical Concept RepresentationThe source studies REFINE, which builds patient-specific temporal graphs from a global medical knowledge graph, selects per-code expansion budgets, and combines heterogeneous graph representations with frozen-LLM semantic refinement. Evidence comes from experiments on MIMIC-III and MIMIC-IV, including ablations, knowledge-graph selection, and data-insufficiency settings.research paper · Sep 3, 2026
Related questions
How can synthetic medical time-series generation preserve rare-class patterns across temporal scales for downstream prediction?How can wearable foundation models organize representations for targeted longitudinal women’s-health prediction?When does clinical text materially influence pixel-level predictions in medical image segmentation?How can reward-based fine-tuning preserve fine-grained fidelity and semantic consistency in conditional medical images?