Guidelines for clinical trial protocols for interventions involving artificial intelligence: the SPIRIT-AI Extension
SPIRIT-AI helps make an AI trial’s intended methods explicit before results are known.
What the paper explores
The SPIRIT-AI extension was developed as a companion to CONSORT-AI, focusing on trial protocols. Its consensus-based items address the AI intervention, deployment setting, users, input and output handling, human interaction and planned analysis of error cases.
How to interpret the evidence
Protocol reporting does not establish that a trial was conducted as planned or that an intervention is effective. The publication is methodological guidance in a journal, not a device authorization.
Questions for MedTech teams
Make operational assumptions and responsibilities visible in the protocol. Clear documentation allows changes and departures from the planned evaluation to be identified later.
Sources & further reading
Read original paperDOI: 10.1136/bmj.m3210
Bibliographic record and abstract on PubMedThis is a short editorial reading note on external research. The original research is attributed to its authors and does not represent an HMC study, product validation or regulatory determination.
