Utilizing ChatGPT to select literature for meta-analysis shows workload reduction while maintaining a similar recall level as manual curation

Abstract

Abstract Background Large language models (LLMs) like ChatGPT showed great potential in aiding medical research. A heavy workload in filtering records is needed during the research process of evidence-based medicine, especially meta-analysis. However, no study tried to use LLMs to help screen records in meta-analysis. Objective In this research, we aimed to explore the possibility of incorporating ChatGPT to facilitate the screening step based on the title and abstract of records during meta-analysis. Methods To assess our strategy, we selected three meta-analyses from the literature, together with a glioma meta-analysis embedded in the study, as additional validation. For the automatic selection of records from curated meta-analyses, a four-step strategy called LARS-GPT was developed, consisting of (1) criteria selection and single-prompt (prompt with one criterion) creation, (2) best combination identification, (3) combined-prompt (prompt with one or more criteria) creation, and (4) request sending and answer summary. Recall, workload reduction, precision, and F1 score were calculated to assess the performance of LARS-GPT. Results A variable performance was found between different single-prompts with a mean recall of 0.841. Based on these single-prompts, we were able to find combinations with performance better than the pre-set threshold. Finally, with a best combination of criteria identified, LARS-GPT showed a 39.5% workload reduction on average with a recall greater than 0.9. Conclusions We show here the groundbreaking finding that automatic selection of literature for meta-analysis is possible with ChatGPT. We provide it here as a pipeline, LARS-GPT, which showed a great workload reduction while maintaining a pre-set recall.

Ähnliche Arbeiten