Clinical Report: Risk Assessment for In-Hospital Mortality in ICU Patients

Overview

This study developed machine learning models to predict in-hospital mortality in ICU patients with cardiovascular disease and diabetes. The models were validated using two large datasets, demonstrating significant accuracy in risk stratification.

Background

Cardiovascular disease and diabetes are prevalent comorbidities that significantly increase mortality risk in hospitalized patients. The coexistence of these conditions complicates clinical management, particularly in the ICU setting. Traditional risk assessment tools may not adequately address the unique challenges posed by this patient population, highlighting the need for advanced predictive models.

Data Highlights

The study utilized data from the MIMIC-IV and eICU Collaborative Research Database to develop and validate machine learning models for mortality prediction.

Key Findings

• Machine learning models were constructed using eight distinct algorithms to predict in-hospital mortality.
• Internal and external validation demonstrated the models' robustness and accuracy.
• SHAP analysis provided insights into the contribution of various clinical variables to mortality risk.
• The developed model serves as a web-based clinical decision calculator for healthcare providers.
• Patients with diabetes and cardiovascular disease in the ICU exhibit a significantly higher risk of in-hospital mortality.

Clinical Implications

The machine learning models developed in this study can assist clinicians in identifying high-risk patients in the ICU, enabling timely interventions. The use of a web-based decision calculator may enhance clinical decision-making and improve patient outcomes.

Conclusion

This research underscores the potential of machine learning in enhancing risk assessment for critically ill patients with cardiovascular disease and diabetes. Further implementation of these models could lead to improved mortality outcomes in this vulnerable population.

Related Resources & Content

1. Frontiers in Endocrinology, 2026 -- Predicting mortality in non-traumatic intracerebral hemorrhage with glucose and lipid data
2. npj Digital Medicine, 2025 -- Unlocking the potential of real-time ICU mortality prediction: redefining risk assessment with continuous data recovery
3. Frontiers in Medicine, 2026 -- Analysis of Risk Factors for Heart Failure in Patients with Type 2 Diabetes Mellitus and Acute ST-Segment Elevation Myocardial Infarction after Percutaneous Coronary Intervention
4. European Journal of Preventive Cardiology -- External Assessment of Cardiovascular Risk Assessment Models in Type 2 Diabetes Patients Utilizing the CARDIANA Cohort from Spain
5. ACC, AHA Issue New Acute Coronary Syndromes Guideline - American College of Cardiology
6. Predictive performance of stress hyperglycemia ratio for poor prognosis in critically ill patients: a systematic review and dose–response meta-analysis | European Journal of Medical Research
7. The Prognostic Performance of Artificial Intelligence and Machine Learning Models for Mortality Prediction in Intensive Care Units: A Systematic Review - PubMed
8. ACC, AHA Issue New Acute Coronary Syndromes Guideline - American College of Cardiology
9. Predictive performance of stress hyperglycemia ratio for poor prognosis in critically ill patients: a systematic review and dose–response meta-analysis | European Journal of Medical Research | Full Text
10. The Prognostic Performance of Artificial Intelligence and Machine Learning Models for Mortality Prediction in Intensive Care Units: A Systematic Review - PubMed