analiza aiqbench
ChatHealthAI addresses the critical gap between structured EHR data and unstructured LLM reasoning by aligning pretrained foundation models with frozen language models via a task-aware resampler. This approach successfully enhances interpretability and reasoning quality in clinical decision support while maintaining competitive predictive performance on standard benchmarks. The work demonstrates a viable pathway for integrating specialized medical data representations with general-purpose LLMs to achieve grounded, explainable clinical predictions.
ChatHealthAI: Aligning Electronic Health Record Representations with Large Language Models for Grounded Clinical Reasoning
arXiv:2606.02802v2 Announce Type: replace Abstract: Large language models (LLMs) exhibit strong natural-language reasoning abilities for clinical decision support, but struggle to effectively model structured longitudinal electronic health records (EHRs). In contrast, EHR foundation models can learn predictive patient representations, yet lack interpretable language-based reasoning. To bridge this gap, we propose ChatHealthAI, a multimodal reasoning framework that aligns structured EHR representations from a pretrained EHR foundation model with the semantic space of a frozen LLM through a task-aware resampler. By integrating longitudinal patient representations with refined clinical event descriptions, ChatHealthAI enables clinically grounded natural-language reasoning while maintaining accurate patient prediction. We evaluated ChatHealthAI on three clinical predictive tasks from the EHRSHOT benchmark. Results show that ChatHealthAI improves reasoning quality and interpretability while preserving competitive predictive performance. These findings highlight the potential of integrating EHR foundation models with pretrained LLMs for interpretable clinical prediction.
Abstract:Large language models (LLMs) exhibit strong natural-language reasoning abilities for clinical decision support, but struggle to effectively model structured longitudinal electronic health records (EHRs). In contrast, EHR foundation models can learn predictive patient representations, yet lack interpretable language-based reasoning. To bridge this gap, we propose ChatHealthAI, a multimodal reasoning framework that aligns structured EHR representations from a pretrained EHR foundation model with the semantic space of a frozen LLM through a task-aware resampler. By integrating longitudinal patient representations with refined clinical event descriptions, ChatHealthAI enables clinically grounded natural-language reasoning while maintaining accurate patient prediction. We evaluated ChatHealthAI on three clinical predictive tasks from the EHRSHOT benchmark. Results show that ChatHealthAI improves reasoning quality and interpretability while preserving competitive predictive performance. These findings highlight the potential of integrating EHR foundation models with pretrained LLMs for interpretable clinical prediction.
Artificial Intelligence (cs.AI)
arXiv:2606.02802 [cs.AI]
(or arXiv:2606.02802v2 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2606.02802 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 229 cuvinte
Subiecte conexe
ChatHealthAI: Aligning Electronic Health Record Representations with Large Language Models for Grounded Clinical Reasoning
arXiv · 3 iun. 2026
CAP: Towards PPG Universal Representation Learning with Patient-level Supervision
arXiv · 16 iun. 2026
LLM4EHR: Aligning Clinical Time Series with Medical Event Sequences via Large Language Models
arXiv · 20 iul. 2026
PPG-LM: A Photoplethysmography-Language Model with Multi-Level Clinical Alignment
arXiv · 29 sept. 2026