What question did this study set out to answer?

This research aims to enhance patient similarity computation through an advanced learning network that utilizes LLMs and EMR data.

May 10, 2026

LELN: A Large Language Model-Dynamically Enhanced Learning Network for Patient Similarity Calculation

Key Points

This research aims to enhance patient similarity computation through an advanced learning network that utilizes LLMs and EMR data.
Developed LELN to dynamically structure EMR data utilizing large language models,
Implemented a dual-stage feature aggregation strategy combining graph attention networks and BiLSTM with attention,
Conducted experiments on real-world Chinese dataset and public MIMIC-III data to evaluate performance.
LELN achieved an F1 score of 87.66% on a real-world Chinese dataset,
LELN also recorded an F1 score of 85.95% on the MIMIC-III dataset,
Experimental results demonstrate LELN's robustness and accuracy compared to existing methods.

Abstract

The rapid expansion of Electronic Medical Record (EMR) data has advanced AI-driven patient similarity computation, a key technology for intelligent healthcare. However, the handling of heterogeneous EMR formats and the integration of domain knowledge constrain existing methods. While graph-based approaches show promise, they still struggle with these issues. To address this, we propose a Large Language Model-Dynamically Enhanced Learning Network (LELN), leveraging LLMs' commonsense knowledge and reasoning to dynamically structure EMR data and enhance medical knowledge integration. LELN in tegrates two LLM-basedmodules:DS-EE(DeepSeek-Event Extraction) extracts medical events to construct structured EMR event graphs, and DS-KB (DeepSeek-Knowledge Base) infers disease-relevant knowledge to augment feature representations. The model employs a dual-stage spatial-temporal feature aggregation strategy: a Graph Attention Network captures intra- and inter-event dependencies, followed by a Bidirectional Long-Short Term Memory (BiLSTM) with attention to model temporal disease progression. Additionally, a clinical prior-guided attention mechanism emphasizes discriminative diagnostic features, improving clinical relevance. Extensive experiments on heterogeneous datasets-a real-world Chinese dataset and public MIMIC-III-show LELN outperforms baselines, achieving F1 scores of 87.66% and 85.95%, demonstrating robustness and accuracy.

Bookmark

Cite This Study

Zhu et al. (Thu,) studied this question.

synapsesocial.com/papers/6a001ff2c8f74e3340f9b2f0 https://doi.org/https://doi.org/10.1109/jbhi.2026.3691375

Bookmark