Crossmark

Development and multi-database validation of interpretable machine learning models for predicting In-Hospital mortality in pneumonia patients: A comprehensive analysis across four healthcare systems

Crossref DOI link: https://doi.org/10.1186/s12931-025-03348-w

Published Online: 2025-09-30

Update policy: https://doi.org/10.1007/springer_crossmark_policy

Authors

Chen, Jiahuan

Hou, Dongni

Song, Yuanlin
License Information

Text and Data Mining valid from 2025-09-30

Version of Record valid from 2025-09-30
More Information

Article History

Received: 25 July 2025

Accepted: 18 August 2025

First Online: 30 September 2025

Declarations

:

: This study utilized both publicly available databases and institutional data. For the publicly available databases (MIMIC-IV, MIMIC-III CareVue, and eICU), ethics approval was waived as these databases contain de-identified patient data that are freely accessible to researchers who have completed the required training courses. For the FAHZU database, this study was approved by the Ethics Committee of Zhongshan Hospital, Fudan University (approval number: B2022-263). Written informed consent was obtained from all participants or their legally authorized representatives prior to enrollment in the prospective multicenter study. The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. All data were handled in compliance with applicable privacy regulations and institutional policies to protect patient confidentiality.

: Not applicable.

: The authors declare no competing interests.

Document is current

Any future updates will be listed below