About the journal

We publish studies of AI in healthcare that matter for patient care: development and validation of clinical models, external and prospective validation, deployment and monitoring, safety and failure analyses, evaluations of clinical large language models, and work on reproducibility and reporting standards.

Negative results and replications are welcome. A careful study showing that a model does not transfer to a new setting is as useful to clinicians as one showing that it does.

Out of scope: AI methods papers with no clinical evaluation, and studies with no AI component.

• Original research
• Brief report
• Systematic review
• Perspective

Model development and validation studies, trials, evaluations of language models, negative results and replications are all original research. We read the design from the paper and apply the matching reporting standard.

Every manuscript first receives a full, evidence-linked review from our AI review system, usually within minutes. Accepted papers are published within 7 days. The decision it proposes is checked by people: two independent experts confirm every acceptance, and a human editor answers every appeal. The full process is described in our peer review policy.

The AI reviews a blinded copy. It never sees author names, institutions or countries.

All articles are free to read on publication. Authors keep their copyright and publish under the CC BY 4.0 licence, which lets anyone share and reuse the work with credit.

Clinical AI Evidence is published by Yesil Science. An ISSN will be requested once the first articles are published.

We follow the COPE Core Practices and will apply to the Directory of Open Access Journals once we have a first volume of published articles.

The editorial office address will be published here before launch. Until then, authors can reach an editor through the appeal button on their submission page.