HypEHR: Hyperbolic Modeling of Electronic Health Records for Efficient Question Answering

arXiv:2604.21027v1 Announce Type: new Abstract: Electronic health record (EHR) question answering is often handled by LLM-based pipelines that are costly to deploy and do not explicitly leverage the hierarchical structure of clinical data. Motivated by evidence that medical ontologies and patient trajectories exhibit hyperbolic geometry, we propose HypEHR, a compact Lorentzian model that embeds codes, visits, and questions in hyperbolic space and answers queries via geometry-consistent cross-attention with type-specific pointer heads. HypEHR is pretrained with next-visit diagnosis prediction and hierarchy-aware regularization to align representations with the ICD ontology. On two MIMIC-IV-based EHR-QA benchmarks, HypEHR approaches LLM-based methods while using far fewer parameters. Our code is publicly available at https://github.com/yuyuliu11037/HypEHR.

Related

• CBR-to-SQL: Rethinking Retrieval-based Text-to-SQL using Case-based Reasoning in the Healthcare Domain
• Explainability in Generative Medical Diffusion Models: A Faithfulness-Based Analysis on MRI Synthesis
• Surrogate modeling for interpreting black-box LLMs in medical predictions
• A Multiparty Homomorphic Encryption Approach to Confidential Federated Kaplan Meier Survival Analysis

Source: arXiv cs.AI | 2026-04-24