EMRModel: A Large Language Model for Extracting Medical Consultation Dialogues into Structured Medical Records

TL;DR

EMRModel targets converting unstructured medical consultation dialogues into structured EMRs. It combines LoRA-based fine-tuning with code-style prompts to produce semi-structured code outputs that are decoded into EMRs, supported by a new high-quality dataset and a fine-grained benchmark. In experiments, EMRModel achieves an $F1$ of $88.1\%$, about a $49.5\%$ improvement over standard pre-trained models, and outperforms traditional LoRA tuning methods. The work enables scalable EMR generation across institutions and lays groundwork for integration with medical knowledge bases and downstream decision-support tasks.

Abstract

Medical consultation dialogues contain critical clinical information, yet their unstructured nature hinders effective utilization in diagnosis and treatment. Traditional methods, relying on rule-based or shallow machine learning techniques, struggle to capture deep and implicit semantics. Recently, large pre-trained language models and Low-Rank Adaptation (LoRA), a lightweight fine-tuning method, have shown promise for structured information extraction. We propose EMRModel, a novel approach that integrates LoRA-based fine-tuning with code-style prompt design, aiming to efficiently convert medical consultation dialogues into structured electronic medical records (EMRs). Additionally, we construct a high-quality, realistically grounded dataset of medical consultation dialogues with detailed annotations. Furthermore, we introduce a fine-grained evaluation benchmark for medical consultation information extraction and provide a systematic evaluation methodology, advancing the optimization of medical natural language processing (NLP) models. Experimental results show EMRModel achieves an F1 score of 88.1%, improving by49.5% over standard pre-trained models. Compared to traditional LoRA fine-tuning methods, our model shows superior performance, highlighting its effectiveness in structured medical record extraction tasks.

Figures (9)

• Figure 1: Example of converting medical consultation dialogues into structured data.
• Figure 2: Study design overview.
• Figure 3: A comparison diagram between the traditional natural language prompt structure (left) and the code-style prompt structure (right).
• Figure 4: Dataset data example.
• Figure 5: Performance comparison of different models and fine-tuning strategies in the task of extracting information from medical consultation dialogues.