Related Experiment Video
Updated: Oct 8, 2025
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Early prediction of clinical deterioration using data-driven machine-learning modeling of electronic health records
Victor M Ruiz1, Michael P Goldsmith2, Lingyun Shi1
1Tsui Laboratory, Department of Biomedical and Health Informatics, Children's Hospital of Philadelphia, Philadelphia, Pa.
Insights
A new machine learning model, the Intensive care Warning Index (I-WIN), accurately predicts clinical deterioration in infants with congenital heart disease up to 8 hours in advance. This data-driven approach offers a potential paradigm shift for early intervention in critical care settings.
Area of Science:
- Critical Care Medicine
- Pediatric Cardiology
- Health Informatics
Background:
- Infants with single-ventricle and shunt-dependent congenital heart disease are at high risk of clinical deterioration.
- Routinely collected electronic health record (EHR) data offers a rich source for predictive modeling.
- Existing risk prediction models may not fully leverage the complexity of EHR data.
Purpose of the Study:
- To develop and evaluate a high-dimensional, data-driven model for predicting clinical deterioration in critically-ill infants.
- To identify high-risk patients using routinely collected EHR data.
Main Methods:
- A retrospective cohort study of 488 infants (<6 months old) with congenital heart disease admitted to the cardiac intensive care unit.
- Development of the Intensive care Warning Index (I-WIN) using machine learning (ensemble of 5 extreme gradient boosting models).
- Systematic assessment of 1028 EHR variables (vital signs, medications, lab tests, diagnoses) for risk prediction.
Main Results:
- The I-WIN model achieved an area under the receiver operating characteristic curve (AUROC) of 0.92 at 4 hours before deterioration.
- High performance was maintained at 8 hours prior to deterioration with an AUROC of 0.815.
- The model demonstrated 0.881 sensitivity and 0.862 specificity in predicting adverse events.
Conclusions:
- The I-WIN model accurately predicts clinical deterioration in critically-ill infants with congenital heart disease up to 8 hours in advance.
- This data-driven, machine-learning approach represents a shift from traditional expert-based risk factor selection.
- The I-WIN model has potential for broader application in data-rich critical care environments.
Objectives:
To develop and evaluate a high-dimensional, data-driven model to identify patients at high risk of clinical deterioration from routinely collected electronic health record (EHR) data.
Materials And Methods:
In this single-center, retrospective cohort study, 488 patients with single-ventricle and shunt-dependent congenital heart disease <6 months old were admitted to the cardiac intensive care unit before stage 2 palliation between 2014 and 2019. Using machine-learning techniques, we developed the Intensive care Warning Index (I-WIN), which systematically assessed 1028 regularly collected EHR variables (vital signs, medications, laboratory tests, and diagnoses) to identify patients in the cardiac intensive care unit at elevated risk of clinical deterioration. An ensemble of 5 extreme gradient boosting models was developed and validated on 203 cases (130 emergent endotracheal intubations, 34 cardiac arrests requiring cardiopulmonary resuscitation, 10 extracorporeal membrane oxygenation cannulations, and 29 cardiac arrests requiring cardiopulmonary resuscitation onto extracorporeal membrane oxygenation) and 378 control periods from 446 patients.
Results:
At 4 hours before deterioration, the model achieved an area under the receiver operating characteristic curve of 0.92 (95% confidence interval, 0.84-0.98), 0.881 sensitivity, 0.776 positive predictive value, 0.862 specificity, and 0.571 Brier skill score. Performance remained high at 8 hours before deterioration with 0.815 (0.688-0.921) area under the receiver operating characteristic curve.
Conclusions:
I-WIN accurately predicted deterioration events in critically-ill infants with high-risk congenital heart disease up to 8 hours before deterioration, potentially allowing clinicians to target interventions. We propose a paradigm shift from conventional expert consensus-based selection of risk factors to a data-driven, machine-learning methodology for risk prediction. With the increased availability of data capture in EHRs, I-WIN can be extended to broader applications in data-rich environments in critical care.
Related Concept Videos
01:18Steps in Outbreak Investigation
01:30Methods of Documentation VII: EMR