Related Experiment Video
Updated: Nov 29, 2025
Description of a Swine Infant Model of Volume-Controlled Hemorrhagic Shock
Published on: November 3, 2023
Decision-making in pediatric blunt solid organ injury: A deep learning approach to predict massive transfusion, need
Niti Shahi1, Ashwani K Shahi2, Ryan Phillips3
1Division of Pediatric Surgery, Children's Hospital Colorado, Aurora, CO, USA; Department of Surgery, University of Colorado School of Medicine, Aurora, CO, USA; Department of Surgery, University of Massachusetts, Worcester, MA, USA.
Insights
Machine learning models accurately predict outcomes for pediatric blunt solid organ injury (BSOI). These algorithms identify patients needing massive transfusion or intervention early, improving care for children with BSOI.
Area of Science:
- Pediatric Trauma Surgery
- Computational Medicine
- Medical Informatics
Background:
- Pediatric blunt solid organ injury (BSOI) management is complex, with declining hemoglobin and hemodynamic instability as key intervention triggers.
- Current clinical management lacks predictive precision for escalating care needs.
- Machine learning (ML) algorithms were explored to identify novel predictors for critical events in pediatric BSOI.
Purpose of the Study:
- To develop and evaluate ML models for predicting critical outcomes in pediatric BSOI within 4 hours of emergency department presentation.
- To identify combinations of clinical, laboratory, and imaging variables that herald the need for massive transfusion (MT), failure of non-operative management (NOM), or mortality.
- To predict successful NOM without intervention.
Main Methods:
- A retrospective analysis of 477 pediatric patients (≤18 years) with BSOI from 2009-2018 at a level 1 trauma center.
- Deep learning models were trained using pre-hospital and ED data, including vital signs, shock index-pediatric adjusted (SIPA), injury grade, laboratory results (hemoglobin, lactate, TEG), and imaging (FAST, CT).
- Model performance was assessed using sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC).
Main Results:
- The study included 477 patients: 5.7% required MT, 7.2% failed NOM, 4.4% died, and 89.1% had successful NOM.
- The developed ML models demonstrated high accuracy: MT (90.5%), failure of NOM (83.8%), mortality (91.9%), and successful NOM (90.3%) in the validation set.
- Serial vital signs, injury grade, hemoglobin, and positive FAST scans showed low correlations with outcomes, highlighting the predictive power of combined features.
Conclusions:
- Deep learning models integrating clinical, laboratory, and radiographic data can accurately predict the need for emergent intervention and mortality in pediatric BSOI within 4 hours of admission.
- These models show high accuracy and sensitivity, offering a promising framework for early identification of high-risk patients.
- External validation and prospective application studies are necessary to confirm the feasibility and clinical utility of these ML-based predictive tools.
Background:
The principal triggers for intervention in the setting of pediatric blunt solid organ injury (BSOI) are declining hemoglobin values and hemodynamic instability. The clinical management of BSOI is, however, complex. We therefore hypothesized that state-of-art machine learning (computer-based) algorithms could be leveraged to discover new combinations of clinical variables that might herald the need for an escalation in care. We developed algorithms to predict the need for massive transfusion (MT), failure of non-operative management (NOM), mortality, and successful non-operative management without intervention, all within 4 hours of emergency department (ED) presentation.
Methods:
Children (≤18 years) who sustained a BSOI (liver, spleen, and/or kidney) between 2009 and 2018 were identified in the trauma registry at a pediatric level 1 trauma center. Deep learning models were developed using clinical values [vital signs, shock index-pediatric adjusted (SIPA), organ injured, and blood products received], laboratory results [hemoglobin, base deficit, INR, lactate, thromboelastography (TEG)], and imaging findings [focused assessment with sonography in trauma (FAST) and grade of injury on computed tomography scan] from pre-hospital to ED settings for prediction of MT, failure of NOM, mortality, and successful NOM without intervention. Sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were used to evaluate each model's performance.
Results:
A total of 477 patients were included, of which 5.7% required MT (27/477), 7.2% failed NOM (34/477), 4.4% died (21/477), and 89.1% had successful NOM (425/477). The accuracy of the models in the validation set was as follows: MT (90.5%), failure of NOM (83.8%), mortality (91.9%), and successful NOM without intervention (90.3%). Serial vital signs, the grade of organ injury, hemoglobin, and positive FAST had low correlations with outcomes.
Conclusion:
Deep learning-based models using a combination of clinical, laboratory and radiographic features can predict the need for emergent intervention (MT, angioembolization, or operative management) and mortality with high accuracy and sensitivity using data available in the first 4 hours of admission. Further research is needed to externally validate and determine the feasibility of prospectively applying this framework to improve care and outcomes.
Level Of Evidence:
III STUDY TYPE: Retrospective comparative study (Prognosis/Care Management).
Related Concept Videos
01:26Kidney Transplant II: Surgical Procedure
01:28Kidney Transplant I: Introduction