Sepsis-AKI Predictors with CD177 and IL18R1 Biomarkers

188 septic patient records, including 89 with sepsis-associated acute kidney injury, integrate clinical variables with transcriptomic-guided blood biomarkers. Weiqin Wu developed this dataset to build a multimodal predictive model, with data last updated in April 2026. It features predictors like SOFA score, MAP, BUN, CRP, and the immune biomarkers CD177 and IL18R1.

Use Cases

• Predict SA-AKI diagnosis using SOFA score, MAP, BUN, CRP, and CD177 gene expression levels via logistic regression.
• Validate the prognostic power of the IL18R1 biomarker against clinical outcomes in an external cohort of 83 patients.
• Perform feature selection with LASSO regression on clinical predictors like mechanical ventilation status and C-reactive protein.
• Benchmark new predictive models against the reported AUC of 0.989 from the combined clinical and biomarker model.

Strengths

• Model demonstrated excellent discrimination with an AUC of 0.989 in internal testing.
• Dataset includes an external validation cohort of 83 patients, where the model maintained an AUC of 0.909.
• Combines 5 key clinical predictors with 2 specific immune-activation biomarkers (CD177, IL18R1) identified from GEO database analysis.

Limitations

• Limited to 188 patient records, which is a small sample size for robust machine learning.
• Retrospective study design from a single center introduces potential selection bias.
• Dataset size is only 157.9 KB, indicating limited raw data or primary feature scope.

Provenance

Source

Retrospective cohort of septic patients from an ICU, with gene expression data validated via qPCR.

Collection Method

Clinical data collection and qPCR quantification of blood samples within 24 hours of ICU admission.

Freshness

Data was last updated in April 2026.