Related Experiment Video
Updated: Nov 13, 2025
Constructing and Visualizing Models using Mime-based Machine-learning Framework
Published on: July 22, 2025
Continuous Prediction of Mortality in the PICU: A Recurrent Neural Network Model in a Single-Center Dataset
Melissa D Aczon1,2, David R Ledbetter1,2, Eugene Laksana1,2
1Department of Anesthesiology and Critical Care Medicine, Children's Hospital Los Angeles, Los Angeles, CA.
Insights
A new recurrent neural network model accurately predicts pediatric intensive care unit mortality risk using electronic health records. This AI tool offers continuous, real-time risk assessment for critically ill children.
Area of Science:
- Artificial Intelligence in Medicine
- Pediatric Critical Care
- Machine Learning for Healthcare
Background:
- Assessing illness severity in pediatric intensive care units (PICUs) is crucial for patient management.
- Existing scoring systems have limitations in providing continuous, real-time risk evaluation.
- Electronic medical records (EMRs) contain rich data for developing advanced predictive models.
Purpose of the Study:
- To develop a proof-of-concept recurrent neural network (RNN) model.
- To utilize EMR data for continuous mortality risk assessment in critically ill children.
- To serve as a proxy for illness severity throughout ICU stays.
Main Methods:
- Retrospective cohort study of 12,516 PICU episodes (9,070 children).
- Development of an RNN model using EMR data, partitioned into training, validation, and test sets.
- Comparison of RNN predictions against established scores like PIM-2, PRISM-III, and PELOD-1.
Main Results:
- RNN achieved an area under the receiver operating characteristic curve (AUC) of 0.94 at 12 hours, outperforming PIM-2 (0.88), PRISM-III (0.89), and PELOD-1 (0.85).
- RNN's predictive performance improved with more data and shorter prediction lead times, reaching an AUC of 0.99 at 24 hours pre-discharge.
- The model demonstrated robust performance across various diagnostic categories and outperformed daily PELOD scores for longer ICU stays.
Conclusions:
- The RNN model effectively processes dynamic EMR data for continuous risk assessment.
- High discrimination suggests the RNN's potential for accurate, real-time evaluation of pediatric ICU patient severity.
- This AI-driven approach offers a promising advancement in monitoring critically ill children.
Objectives:
Develop, as a proof of concept, a recurrent neural network model using electronic medical records data capable of continuously assessing an individual child's risk of mortality throughout their ICU stay as a proxy measure of severity of illness.
Design:
Retrospective cohort study.
Setting:
PICU in a tertiary care academic children's hospital.
Patients/Subjects:
Twelve thousand five hundred sixteen episodes (9,070 children) admitted to the PICU between January 2010 and February 2019, partitioned into training (50%), validation (25%), and test (25%) sets.
Interventions:
None.
Measurements And Main Results:
On 2,475 test set episodes lasting greater than or equal to 24 hours in the PICU, the area under the receiver operating characteristic curve of the recurrent neural network's 12th hour predictions was 0.94 (CI, 0.93-0.95), higher than those of Pediatric Index of Mortality 2 (0.88; CI, [0.85-0.91]; p < 0.02), Pediatric Risk of Mortality III (12th hr) (0.89; CI, [0.86-0.92]; p < 0.05), and Pediatric Logistic Organ Dysfunction day 1 (0.85; [0.81-0.89]; p < 0.002). The recurrent neural network's discrimination increased with more acquired data and smaller lead time, achieving a 0.99 area under the receiver operating characteristic curve 24 hours prior to discharge. Despite not having diagnostic information, the recurrent neural network performed well across different primary diagnostic categories, generally achieving higher area under the receiver operating characteristic curve for these groups than the other three scores. On 692 test set episodes lasting greater than or equal to 5 days in the PICU, the recurrent neural network area under the receiver operating characteristic curves significantly outperformed their daily Pediatric Logistic Organ Dysfunction counterparts (p < 0.005).
Conclusions:
The recurrent neural network model can process hundreds of input variables contained in a patient's electronic medical record and integrate them dynamically as measurements become available. Its high discrimination suggests the recurrent neural network's potential to provide an accurate, continuous, and real-time assessment of a child in the ICU.
