Scholarlink

This paper presents a study on the application of Large Language Models (LLMs) for the automatic classification of ischemic stroke subtypes from Electronic Health Records (EHRs). The study involves the creation and analysis of four datasets, an evaluation of model performance using standard classification metrics, and a comparison with expert manual labeling. Results suggest LLMs can offer promising outcomes even under computational constraints, paving the way for clinical automation and diagnostic support.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Automated Classification of Ischemic Stroke Subtypes from Electronic Health Records Using Large Language Models

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Alejandro Vásquez,
  • Cristina Rubio-Escudero,
  • Germán Antonio Escobar-Rodríguez

摘要

This paper presents a study on the application of Large Language Models (LLMs) for the automatic classification of ischemic stroke subtypes from Electronic Health Records (EHRs). The study involves the creation and analysis of four datasets, an evaluation of model performance using standard classification metrics, and a comparison with expert manual labeling. Results suggest LLMs can offer promising outcomes even under computational constraints, paving the way for clinical automation and diagnostic support.