An interpretable machine learning algorithm enables dynamic 48-hour mortality prediction during an ICU stay
LGBM-48h provides a dynamic and interpretable framework for short-term ICU mortality prediction. The model may support clinical decision-making and prioritization of care, but requires further validation in real-time and prospective settings.
Authors
- Simon Lindner (ORCID: https://orcid.org/0000-0003-2424-2066)
- M. Britsch (ORCID: https://orcid.org/0009-0004-7257-6057)
- Verena Schneider‐Lindner (ORCID: https://orcid.org/0000-0001-5207-8203)
- Simone Britsch (ORCID: https://orcid.org/0009-0001-2212-4114)
- Daniel Duerschmied (ORCID: https://orcid.org/0000-0001-5249-4012)
- Tobias Becher (ORCID: https://orcid.org/0000-0001-7670-757X)
- Thomas Helbing (ORCID: https://orcid.org/0009-0009-9210-9559)
- Leonie Hahn
- Manfred Thiel
Institutions
- Heidelberg University (DE)
- University Hospital Heidelberg (DE)
- University Medical Centre Mannheim (DE)
Publication Details
- Journal
- Communications Medicine
- Published
- 2025-10-15
- DOI
- https://doi.org/10.1038/s43856-025-01192-z
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00