Clinical Report: Predicting 30-Day Mortality in Immunocompromised Sepsis Patients

Overview

This study developed a machine learning model to predict 30-day mortality in immunocompromised sepsis patients using clinical data collected within 6 hours of ICU admission. The Support Vector Machine (SVM) model demonstrated an AUC of 0.794 in the validation set and 0.847 in external validation.

Background

Sepsis is a leading cause of death in ICU patients, particularly affecting immunocompromised individuals who exhibit different pathophysiological responses compared to non-immunosuppressed patients. The urgency for effective risk stratification tools is heightened by the unique challenges faced by this vulnerable group, as traditional prognostic scoring systems may not adequately predict outcomes. Early identification of high-risk patients can facilitate timely interventions and resource allocation.

Data Highlights

Key Findings

• A total of 2,494 immunosuppressed sepsis patients were included in the study.
• The final prediction model utilized 10 clinical feature variables.
• The SVM model outperformed other machine learning algorithms tested.
• Calibration curves indicated good consistency between predicted probabilities and actual risk.
• External validation demonstrated strong generalizability of the model.

Clinical Implications

The SVM-based prediction model provides a tool for identifying high-risk immunocompromised sepsis patients based on clinical indicators available within the first 6 hours of ICU admission.

Conclusion

This study developed an SVM-based prediction model for 30-day mortality risk in immunocompromised sepsis patients using clinical indicators available within 6 hours of admission.

Related Resources & Content

1. Frontiers in Medicine, 2026 -- Machine Learning Model for Predicting the Risk of AKI in Early Hemodynamically Stable Sepsis Patients
2. DIGITAL HEALTH, 2026 -- Incremental domain adaptation-based ICU patient mortality prediction
3. Frontiers in Cardiovascular Medicine, 2026 -- A machine learning model for predicting short-term in-hospital mortality in acute myocardial infarction with coexisting chronic obstructive pulmonary disease
4. Frontiers in Medicine, 2026 -- Prediction models for mortality in patients with sepsis: a systematic review and meta-analysis
5. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) - PMC
6. Clinical characteristics, risk factors and outcome of critically ill immunocompromised patients with bloodstream infections and sepsis | PLOS One
7. Explainable machine learning model for prediction of 28-day all-cause mortality in immunocompromised patients in the intensive care unit: a retrospective cohort study based on MIMIC-IV database - PubMed
8. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) - PMC
9. Clinical characteristics, risk factors and outcome of critically ill immunocompromised patients with bloodstream infections and sepsis | PLOS One
10. Explainable machine learning model for prediction of 28-day all-cause mortality in immunocompromised patients in the intensive care unit: a retrospective cohort study based on MIMIC-IV database - PubMed