Artikel
Zero-Shot LLMs for Named Entity Recognition: Targeting Cardiac Function Indicators in German Clinical Texts
Suche in Medline nach
Autoren
Veröffentlicht: | 6. September 2024 |
---|
Gliederung
Text
Large Language Models (LLMs) like ChatGPT have become increasingly prevalent. In medicine, many potential areas arise where LLMs may offer added value. Our research focuses on the use of open-source LLM alternatives like Llama 3, Gemma, Mistral, and Mixtral to extract medical parameters from German clinical texts. We concentrate on German due to an observed gap in research for non-English tasks. In detail, we extracted 14 cardiovascular function indicators, including left and right ventricular ejection fraction (LV-EF and RV-EF), from 497 variously formulated cardiac magnetic resonance imaging (MRI) reports. Our systematic analysis confirms strong performance with up to 95.4 % right annotation (99.8 % named entity recognition (NER) accuracy) across different architectures, despite the fact that these models were not explicitly fine-tuned for data extraction and German language. This results in a strong recommendation to use open-source LLMs for extracting medical parameters from clinical texts, including those in German.
The authors declare that they have no competing interests.
The authors declare that a positive ethics committee vote has been obtained.