The use of artificial intelligence in dermatology systematic reviews: A comparative analysis of Elicit against human reviewers

Vyas, Jui ORCID: https://orcid.org/0000-0003-2839-2651, Johns, Jeffrey R., Forrest, Emily, Hilliar, Mari Ann, Salek, Sam and Finlay, Andrew Y. ORCID: https://orcid.org/0000-0003-2143-1646 2026. The use of artificial intelligence in dermatology systematic reviews: A comparative analysis of Elicit against human reviewers. Acta Dermato-Venereologica 106 , adv–2025-0213. 10.2340/actadv.v106.adv-2025-0213

Preview

PDF - Published Version Available under License Creative Commons Attribution Non-commercial. Download (624kB) | Preview

Abstract

Systematic reviews (SRs) are crucial to support evidence-based medicine but are highly labour-intensive to conduct given the need to review all available literature, extract relevant data, analyse results and form conclusions (1, 2). Elicit, a journal article discovery tool powered by artificial intelligence (AI), is designed specifically to streamline the process of literature review, evidence synthesis and data extraction from scholarly sources; this could potentially streamline initial screening and data extraction in SRs. Elicit allows the use of “natural language questions” instead of keywords to quickly find and analyse articles (3). Our aim was to compare the efficiency and accuracy of Elicit against a published manual search and extraction SR (4) of the Dermatology Life Quality Index (DLQI) used as the primary outcome in clinical trials. Two aspects were examined: article searching and data extraction.

Item Type:	Article
Date Type:	Published Online
Status:	Published
Schools:	Schools > Medicine
Publisher:	Medical Journals Sweden
ISSN:	0001-5555
Date of First Compliant Deposit:	16 February 2026
Date of Acceptance:	22 January 2026
Last Modified:	25 Mar 2026 09:49
URI:	https://orca.cardiff.ac.uk/id/eprint/184875

Actions (repository staff only)

Edit Item