STARD-AI Checklist
Interactive reporting checklist for diagnostic accuracy studies that evaluate artificial intelligence and machine learning index tests.
All STARD-AI items
0%
Items Rated
0
Reported
0
Not Reported
0
N/A
0
Unrated
About This Tool: The STARD-AI Checklist
What is it for?
This tool is a web-based checklist implementing the STARD-AI (Standards for Reporting Diagnostic Accuracy – Artificial Intelligence) reporting guideline. Its primary purpose is to serve as a reporting completeness checklist for diagnostic test accuracy studies that evaluate an AI- or machine-learning-based index test. STARD-AI is a single integrated 40-item reporting guideline for AI diagnostic accuracy studies. It incorporates inherited STARD 2015 items, modifies selected STARD 2015 items, and introduces additional AI-specific items. By using this checklist, authors, editors and reviewers can systematically evaluate whether a study transparently reports the elements needed to appraise the bias, applicability and generalisability of an AI diagnostic test.
How the item tags work
Items shown under a STARD 2015 label are inherited from STARD 2015 unchanged. Items shown under a STARD-AI label carry one of two tags: a New item is a requirement introduced by STARD-AI for AI diagnostic accuracy studies, while a Modified item is a STARD 2015 item that has been reworded for the AI setting. Both designations follow the new (*) and modified (†) markings in the STARD-AI publication and are not editorial additions.
This web tool is provided for educational and review purposes only and does not constitute formal editorial, regulatory, methodological, clinical, or legal advice. Please consult the original publications for the full official guidance.
Sources and attribution
Checklist item wording is reproduced from the CC BY accepted manuscript of the STARD-AI reporting guideline, whose spelling and hyphenation it follows. Explanatory summaries have been independently rewritten for educational and review purposes, drawing on the STARD-AI guideline and the STARD 2015 Explanation & Elaboration document for inherited diagnostic accuracy concepts. Licensing terms for each source are noted below.
1. Sounderajah, V., Guni, A., Liu, X. et al. The STARD-AI reporting guideline for diagnostic accuracy studies using artificial intelligence. Nat Med 31, 3283–3289 (2025). doi:10.1038/s41591-025-03953-8. Item wording here is taken from the peer-reviewed accepted manuscript, which contains the full checklist table and is deposited under CC BY 4.0 at the University of Birmingham research portal — reproduction permitted with attribution. Author Correction: Nat Med (2026), doi:10.1038/s41591-026-04570-9 (authorship only; no change to the checklist).
2. Cohen, J.F., Korevaar, D.A., Altman, D.G. et al. STARD 2015 guidelines for reporting diagnostic accuracy studies: explanation and elaboration. BMJ Open 6, e012799 (2016). doi:10.1136/bmjopen-2016-012799. Published under CC BY-NC 4.0. Consulted as background for the inherited STARD 2015 concepts; no text is reproduced from it.
This checklist is not endorsed by the original authors.
What is it for?
This tool is a web-based checklist implementing the STARD-AI (Standards for Reporting Diagnostic Accuracy – Artificial Intelligence) reporting guideline. Its primary purpose is to serve as a reporting completeness checklist for diagnostic test accuracy studies that evaluate an AI- or machine-learning-based index test. STARD-AI is a single integrated 40-item reporting guideline for AI diagnostic accuracy studies. It incorporates inherited STARD 2015 items, modifies selected STARD 2015 items, and introduces additional AI-specific items. By using this checklist, authors, editors and reviewers can systematically evaluate whether a study transparently reports the elements needed to appraise the bias, applicability and generalisability of an AI diagnostic test.
How the item tags work
Items shown under a STARD 2015 label are inherited from STARD 2015 unchanged. Items shown under a STARD-AI label carry one of two tags: a New item is a requirement introduced by STARD-AI for AI diagnostic accuracy studies, while a Modified item is a STARD 2015 item that has been reworded for the AI setting. Both designations follow the new (*) and modified (†) markings in the STARD-AI publication and are not editorial additions.
This web tool is provided for educational and review purposes only and does not constitute formal editorial, regulatory, methodological, clinical, or legal advice. Please consult the original publications for the full official guidance.
Sources and attribution
Checklist item wording is reproduced from the CC BY accepted manuscript of the STARD-AI reporting guideline, whose spelling and hyphenation it follows. Explanatory summaries have been independently rewritten for educational and review purposes, drawing on the STARD-AI guideline and the STARD 2015 Explanation & Elaboration document for inherited diagnostic accuracy concepts. Licensing terms for each source are noted below.
1. Sounderajah, V., Guni, A., Liu, X. et al. The STARD-AI reporting guideline for diagnostic accuracy studies using artificial intelligence. Nat Med 31, 3283–3289 (2025). doi:10.1038/s41591-025-03953-8. Item wording here is taken from the peer-reviewed accepted manuscript, which contains the full checklist table and is deposited under CC BY 4.0 at the University of Birmingham research portal — reproduction permitted with attribution. Author Correction: Nat Med (2026), doi:10.1038/s41591-026-04570-9 (authorship only; no change to the checklist).
2. Cohen, J.F., Korevaar, D.A., Altman, D.G. et al. STARD 2015 guidelines for reporting diagnostic accuracy studies: explanation and elaboration. BMJ Open 6, e012799 (2016). doi:10.1136/bmjopen-2016-012799. Published under CC BY-NC 4.0. Consulted as background for the inherited STARD 2015 concepts; no text is reproduced from it.
This checklist is not endorsed by the original authors.
