Skip to main content

# eICU-CRD

2 related articles

Subscribe to updates

researchsquare.com Oct 22, 2026
Sepsis coagulopathy screening misses its highest-risk patients: 72-hour INR and PTT trajectories recover a group with nearly double the mortality
Sepsis coagulopathy screening still rests on a single admission lab set, with platelets, INR and SOFA combined into one cross-sectional threshold. Among 3,138 sepsis patients in MIMIC-IV, clustering the joint 72-hour trajectories of platelet count, INR and PTT yielded a high-risk subphenotype with 27.8% mortality, of whom the admission SIC score flagged only 18.1%. The gradient reproduced in an independent refit of 12,828 eICU-CRD patients, where 86.3% of the high group remained SIC-negative.
Sepsis coagulopathy screening still rests on a single admission lab set, with platelets, INR and SOFA combined into one cross-sectional threshold. Among 3,138 sepsis patients in MIMIC-IV, clustering the joint 72-hour trajectories of platelet count, INR and PTT yielded a high-risk subphenotype with 27.8% mortality, of whom the admission SIC score flagged only 18.1%. The gradient reproduced in an independent refit of 12,828 eICU-CRD patients, where 86.3% of the high group remained SIC-negative.
Second Affiliated Hospital & Yuying Children's Hospital of Wenzhou Medical University First Affiliated Hospital of Wenzhou Medical University Second Affiliated Hospital of Zhejiang University
Xueshu Yu（First author）
Sepsis-Associated Coagulopathy Gaussian Mixture Model Sepsis-Induced Coagulopathy (SIC) Score Target Trial Emulation MIMIC-IV
researchsquare.com Oct 23, 2026
A five-part data-quality profile flags bad clinical training data, but cannot rank how much harm each defect does
Clinical prediction models inherit their quality from routine-care training data, yet missingness rates and class ratios are not tied to what happens after deployment. An exploratory benchmark across three MIMIC-IV cohorts and the multi-institution eICU database splits data quality into completeness, label noise, distributional similarity, representativeness and temporal integrity, training a real model under each controlled condition and measuring held-out log-loss and calibration change. The composite moved in the expected direction in all three controlled cohorts, but combined corruptions were sub-additive rather than multiplicative, and across 21 analyzable eICU hospitals it tracked calibration but not detectably discrimination.
Clinical prediction models inherit their quality from routine-care training data, yet missingness rates and class ratios are not tied to what happens after deployment. An exploratory benchmark across three MIMIC-IV cohorts and the multi-institution eICU database splits data quality into completeness, label noise, distributional similarity, representativeness and temporal integrity, training a real model under each controlled condition and measuring held-out log-loss and calibration change. The composite moved in the expected direction in all three controlled cohorts, but combined corruptions were sub-additive rather than multiplicative, and across 21 analyzable eICU hospitals it tracked calibration but not detectably discrimination.
Mayo Clinic
Yunguo Yu（First author, Corresponding author）
Data-Quality Profile Clinical Prediction MIMIC-IV eICU-CRD Distribution Shift

You’re all caught up