Interpretable machine learning model for early prediction of 28-day mortality in ICU patients with sepsis-induced coagulopathy: development and validation
Problem
Early prediction of 28-day death risk in ICU patients with sepsis-induced coagulopathy lacked an interpretable, generalizable model.
Why it matters
Sepsis-induced coagulopathy is common and strongly associated with poor outcomes, so early risk stratification matters.
What the AI did
A machine learning model was built with recursive feature elimination and tenfold cross-validation on MIMIC-III, then externally validated on MIMIC-IV and eICU-CRD cohorts.
Results
not stated in source
Impact
External validation on independent databases is described, but no performance metrics are stated in the abstract.
Record
| Venue | European journal of medical research |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Shi Zhou, Zongqing Lu, Yu Liu, Minjie Wang, Wuming Zhou, Xuanxuan Cui, Jin Zhang, Wenyan Xiao et al. |
| Citations | 52 (OpenAlex, at import time) |
| References | 35 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1186/s40001-023-01593-7 |
| Open access | gold |
| Keywords | machine learning, sepsis-induced coagulopathy, mortality prediction, MIMIC-III, interpretable model |
Importance rating
significant — thousands to millions affected
Abstract states sepsis-induced coagulopathy is extremely common in sepsis patients and significantly associated with poor outcomes; early 28-day mortality risk stratification supports ICU decisions.