NMI Biomarker Data for Sepsis Severity and Mortality Prediction

399 adult patients diagnosed with sepsis were recruited to evaluate the predictive value of the serum biomarker N-myc and STAT interactor (NMI). The study used LASSO, logistic regression, and ROC analysis on a training set of 302 patients, with internal validation on 97 patients. The dataset, authored by Xuan Leng and last updated in April 2026, reports NMI concentrations associated with 30-day mortality and septic shock.

Use Cases

• Predicting 30-day mortality in sepsis patients based on serum NMI concentrations.
• Assessing the risk of septic shock based on biomarker levels mentioned in the study.
• Comparing the predictive performance of a novel biomarker against traditional scoring systems like APACHE II and SOFA.
• Building survival analysis models using the Kaplan-Meier survival curves stratified by NMI levels described in the results.

Strengths

• Includes data from 399 adult patients, providing a substantive cohort for analysis.
• Reports specific biomarker concentration values, such as 166.1 for patients with septic shock and 54.2 for sepsis patients.
• Demonstrates high predictive performance with AUC values of 0.86 for mortality and 0.92 for septic shock.
• Employs a rigorous validation approach with a dedicated internal validation set of 97 patients.

Limitations

• Row count and column-level documentation are unknown, which limits suitability assessment and requires manual inspection after download.
• The dataset is small in scale at 286.7 KB, indicating limited scope.
• The primary data file is a DOCX document, which may require parsing to extract structured data.

Provenance

Source

figshare

Collection Method

Prospective research study recruiting patients diagnosed with sepsis.

Freshness

Last updated 2026-04-27 05:16:24; freshness should be verified.