Related Experiment Video
Updated: Sep 21, 2025
Constructing and Visualizing Models using Mime-based Machine-learning Framework
Published on: July 22, 2025
Clinically Interpretable Machine Learning Models for Early Prediction of Mortality in Older Patients with Multiple
Xiaoli Liu1,2,3, Clark DuMontier4,5, Pan Hu6,7
1Key Laboratory for Biomechanics and Mechanobiology of Ministry of Education, Beijing Advanced Innovation Center for Biomedical Engineering, School of Biological Science and Medical Engineering, Beihang University, Beijing, China.
Background:
Multiple organ dysfunction syndrome (MODS) is associated with a high risk of mortality among older patients. Current severity scores are limited in their ability to assist clinicians with triage and management decisions. We aim to develop mortality prediction models for older patients with MODS admitted to the ICU.
Methods:
The study analyzed older patients from 197 hospitals in the United States and 1 hospital in the Netherlands. The cohort was divided into the young-old (65-80 years) and old-old (≥80 years), which were separately used to develop and evaluate models including internal, external, and temporal validation. Demographic characteristics, comorbidities, vital signs, laboratory measurements, and treatments were used as predictors. We used the XGBoost algorithm to train models, and the SHapley Additive exPlanations (SHAP) method to interpret predictions.
Results:
Thirty-four thousand four hundred and ninety-seven young-old (11.3% mortality) and 21 330 old-old (15.7% mortality) patients were analyzed. Discrimination AUROC of internal validation models in 9 046 U.S. patients was as follows: 0.87 and 0.82, respectively; discrimination of external validation models in 1 905 EUR patients was as follows: 0.86 and 0.85, respectively; and discrimination of temporal validation models in 8 690 U.S. patients: 0.85 and 0.78, respectively. These models outperformed standard clinical scores like Sequential Organ Failure Assessment and Acute Physiology Score III. The Glasgow Coma Scale, Charlson Comorbidity Index, and Code Status emerged as top predictors of mortality.
Conclusions:
Our models integrate data spanning physiologic and geriatric-relevant variables that outperform existing scores used in older adults with MODS, which represents a proof of concept of how machine learning can streamline data analysis for busy ICU clinicians to potentially optimize prognostication and decision making.
Insights
New machine learning models accurately predict mortality in older adults with multiple organ dysfunction syndrome (MODS). These models offer improved prognostication for intensive care unit (ICU) patients, outperforming existing clinical scores.
Area of Science:
- Critical Care Medicine
- Geriatric Medicine
- Machine Learning in Healthcare
Background:
- Multiple organ dysfunction syndrome (MODS) presents a significant mortality risk in elderly populations.
- Existing clinical severity scores have limitations in guiding triage and management for older patients with MODS.
- There is a need for improved mortality prediction models tailored to older adults in the ICU.
Purpose of the Study:
- To develop and validate machine learning models for predicting mortality in older patients with MODS admitted to the ICU.
- To compare the performance of developed models against established clinical severity scores.
- To identify key predictors of mortality in this patient cohort.
Main Methods:
- Analysis of a large cohort of older patients (65+ years) from US and Dutch hospitals.
- Development and validation (internal, external, temporal) of models using XGBoost algorithm.
- Utilized demographic data, comorbidities, vital signs, lab results, and treatments as predictors.
- Employed SHapley Additive exPlanations (SHAP) for model interpretability.
Main Results:
- Models demonstrated strong discrimination (AUROC 0.78-0.87) across validation sets for both young-old and old-old groups.
- Developed models significantly outperformed Sequential Organ Failure Assessment (SOFA) and Acute Physiology Score III (APS III).
- Glasgow Coma Scale, Charlson Comorbidity Index, and Code Status were identified as key mortality predictors.
Conclusions:
- Machine learning models integrating physiologic and geriatric variables offer superior prognostication for older adults with MODS.
- These models represent a proof of concept for leveraging AI to enhance clinical decision-making in ICUs.
- The findings suggest potential for optimizing patient management and resource allocation through improved predictive accuracy.
