Predicting hospital admissions, ICU utilization, and prolonged length of stay among febrile pediatric emergency department patients using incomplete and imbalanced electronic health record (EHR) data strategies
Both XGB and LR with median imputation demonstrated robust performance in predicting admissions in the presence of missing data. However, deriving clinically useful models for rare outcomes, such as ICU use or PLOS, remains a challenge due to poor precision/recall and calibration performance. Further research is needed to improve the prediction of rare outcomes in this population.
Authors
- Ioannis Koutroulis (ORCID: https://orcid.org/0000-0002-8396-9022)
- Panagiotis Kratimenos (ORCID: https://orcid.org/0000-0002-7368-0436)
- Andrew R. Clark (ORCID: https://orcid.org/0000-0003-4996-8322)
- Tom Velez (ORCID: https://orcid.org/0000-0003-2157-9165)
- Danielle Velez (ORCID: https://orcid.org/0000-0002-3833-5557)
- Kenneth McKinley (ORCID: https://orcid.org/0000-0003-4624-5778)
- Zara Ibrahim
- Maria Triantafyllou (ORCID: https://orcid.org/0009-0008-6123-4384)
- Kanayo Duru
- Pasha Saif
Institutions
- George Washington University Hospital (US)
- Children's National (US)
- Hasbro Children's Hospital (US)
- Applied Technology Associates (United States) (US)
- Carilion Clinic (US)
- Virginia Tech (US)
Publication Details
- Journal
- International Journal of Medical Informatics
- Published
- 2025-04-06
- DOI
- https://doi.org/10.1016/j.ijmedinf.2025.105905
- Citations
- 2
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 4.96
Funders
- National Institute of Allergy and Infectious Diseases
- Division of Microbiology and Infectious Diseases, National Institute of Allergy and Infectious Diseases