Natural language processing and language models for Dutch clinical text: a systematic review

Abstract

Background. Increasingly natural language processing (NLP) tools and applications - including those using large language models (LLMs) - are developed and used in electronic health records (EHRs). As generalization to specific languages and EHR settings or departments is not guaranteed, having a comprehensive overview about tool availability and accuracy across EHR settings is increasingly important for their effective application and reuse of existing tools. No such overview was available for Dutch, focused on language technology for EHRs that covers the past decade. Objective. To identify and describe existing NLP tools, including those using LLMs, that have been developed or evaluated in real-world Dutch EHRs. Methods. A literature search was conducted in Scopus and Pubmed up to November 11, 2025. Information about the NLP task, patient group, healthcare setting, code and model availability were extracted. Results. A total of 44 studies were included, describing 794 models and 792 evaluations. Most studies focused on information extraction (73%), followed by de-identification (14%), generative applications (11%), and language modeling tasks (9%). Rule-based methods were most frequently used at the study level (50%), while transformer-based approaches accounted for the majority of models and evaluations (55%). Prompting LLMs was used in 16% of studies and accounted for 32% of models and evaluations. Code was shared in 43% of studies, covering 91% of models, whereas only 4% of models were publicly available. Conclusions. A diverse set of NLP models has been developed for Dutch clinical text, with an increasing use of transformer-based and LLM-based approaches. However, model availability remains limited. This review provides a structured overview of available tools and evaluations to support their application and reuse in Dutch clinical settings.

Ähnliche Arbeiten