Dataset of original research that evaluated the quality of generative artificial intelligence-synthesised advice on resuscitation and first aid

Published: 20 June 2025| Version 1 | DOI: 10.17632/dtfptcxxfk.1
Contributor:
Alexei Birkun

Description

This dataset contains the results of an analysis of original studies published in English between 2017 and 2025 that reported quantitative data on the quality (accuracy, correctness, completeness, appropriateness) of generative artificial intelligence-synthesised advice for laypeople on cardiopulmonary resuscitation or first aid. The search for eligible papers was conducted in May 2025 using PubMed, Scopus, and Google Scholar. The dataset table collected the following data: article author(s) and year of publication, publication source link, publication type, study objectives, study design, target health condition(s), name(s) of tested generative artificial intelligence tool(s), a summary of research methods, evaluation metrics, number of evaluations, a summary of study results, conclusions and author recommendations related to the quality of synthesised information. The dataset was utilised to perform the related scoping review.

Files

Categories

Artificial Intelligence, Emergency Medicine, Communication, Internet, Cardiopulmonary Resuscitation, Resuscitation, First Aid, Adult Emergency, Chatbot, Generative Artificial Intelligence, Large Language Model, Generative Pre-Trained Transformer 4

Licence