Please confirm each item before submitting. Incomplete submissions will be returned.

1.    The manuscript fits the Aims & Scope of JMIDH and involves health data, a health problem or a healthcare setting.

2.    The work is original and not under consideration elsewhere; related conference papers and preprints are disclosed and cited.

3.    All authors meet ICMJE criteria and have approved the manuscript.

4.    The title page is a separate file containing author details, ORCID iDs and all declarations.

5.    The manuscript, file properties, code links and supplementary files are anonymised for double-blind review.

6.    The abstract, keywords and word count follow the requirements for the article type.

7.    The relevant reporting guideline checklist is attached.

8.    Ethics approval or a documented exemption is attached; consent or its waiver is described.

9.    Trials of digital interventions were prospectively registered, and the registration number is in the abstract.

10. No identifiable patient information appears in the text, figures, screenshots, code or data.

11. Code availability and data availability statements are included, and the code and data are ready to share or restrictions are explained.

12. For AI studies: data splitting, validation, calibration, subgroup performance and limitations are reported.

13. For studies of large language models: model versions, dates, prompts and settings are reported.

14. Funding and competing interests, including commercial and start-up interests, are disclosed.

15. Use of generative AI in preparing the manuscript is disclosed.

16. References follow Vancouver style and have been verified; datasets and software are cited.