Related Experiment Video
Updated: Jul 11, 2026
A Neonatal Imaging Model of Gram-Negative Bacterial Sepsis
Published on: August 12, 2020
Development and multicenter validation of an explainable machine learning diagnostic criteria for pediatric abdominal
Suqi Cao1, Duote Cai2, Shuhao Zhang2
1National Clinical Research Center for Child And Adolescents' Heath and Diseases, The Children's Hospital, Zhejiang University School of Medicine, Hangzhou, China.
Insights
Early pediatric abdominal sepsis (PAS) is hard to detect. A new explainable AI model, ABSeD, uses routine data to accurately identify PAS in children, improving timely treatment and outcomes.
Area of Science:
- Pediatric critical care medicine
- Medical artificial intelligence
- Clinical decision support systems
Background:
- Pediatric abdominal sepsis (PAS) diagnosis is challenging due to limitations in current criteria.
- Existing tools often overlook early intra-abdominal infections, focusing on cardiopulmonary issues.
Purpose of the Study:
- To develop and validate an explainable machine learning model for early detection of pediatric abdominal sepsis (PAS).
- To create a clinical decision support tool that addresses the gap in identifying intra-abdominal infections.
Main Methods:
- Utilized retrospective data from 6566 pediatric patients (2019-2023) for model construction.
- Employed explainable machine learning algorithms, including hyper-parameter tuning.
- Validated the model prospectively on 308 patients across seven hospitals (Jan-Mar 2025).
Main Results:
- The developed Abdominal Sepsis Diagnosis (ABSeD) model integrates nine routine clinical variables.
- Achieved high diagnostic accuracy on the training set (AUC=0.934, accuracy=0.870, precision=0.910).
- Demonstrated robust multicenter generalizability with strong validation results (AUC=0.928, accuracy=0.873, precision=0.924).
Conclusions:
- The ABSeD model provides an accurate, explainable, and practical tool for early PAS detection.
- This digital tool has the potential to significantly enhance timely interventions for children with intra-abdominal septic pathology.
- The model's performance suggests a valuable addition to clinical decision support for pediatric sepsis.
Abstract:
Accurate identification of early pediatric abdominal sepsis (PAS) is essential to improving outcomes, yet most existing pediatric sepsis criteria and scoring tools primarily focus on cardiopulmonary dysfunction and overlook early intra-abdominal infections. To address this gap, we combined the real-world data with explainable machine learning to develop the Abdominal Sepsis Diagnosis model (ABSeD) for clinical decision support. The model construction used the retrospective data from 6566 pediatric patients who were admitted to the Children's Hospital, Zhejiang University School of Medicine from 2019 to 2023. Prospective data from 308 recruited patients across seven independent hospitals collected between January and March 2025 served as an external validation cohort. PAS status was determined through consensus or by reviewing laparoscopic surgery records. Multiple machine learning algorithms were compared, and the optimal model was further refined by hyper-parameter tuning. The ABSeD model, integrating nine routine clinical variables, demonstrated high diagnostic accuracy (training set: AUC = 0.934, 95% CI: [0.912, 0.950]; accuracy = 0.870, precision = 0.910), and robust multicenter generalizability (AUC = 0.928, 95% CI: [0.895, 0.961]; accuracy = 0.873, precision = 0.924). This model offers an explainable and practical digital tool for early detection of PAS, with potential to enhance timely intervention in hospitalized children with suspected or clinically identified intra-abdominal septic pathology.
