Effective diagnosis of sepsis in critically ill children using probabilistic graphical model
Related Experiment Video
Updated: Jul 30, 2025
A Data-Driven Approach to Quantifying Immune States in Sepsis
Published on: February 7, 2025
Tuong Minh Nguyen1, Kim Leng Poh1, Shu-Ling Chong2,3
1Department of Industrial Engineering and Management, National University of Singapore, Singapore, Singapore.
Insights
Probabilistic graphical models show promise for diagnosing pediatric sepsis in the ICU. Combining vital signs, symptoms, and lab/microbiological tests improved diagnostic accuracy, aiding clinical decision-making.
Area of Science:
- Computational medicine
- Clinical informatics
- Pediatric critical care
Background:
- Probabilistic graphical models (PGMs) offer a robust framework for analyzing complex variable associations.
- Their application in diagnosing pediatric sepsis, particularly in intensive care units (ICUs), is currently limited.
- This study investigates the potential of PGMs to enhance pediatric sepsis diagnosis.
Purpose of the Study:
- To explore the utility of probabilistic graphical models (PGMs) for diagnosing pediatric sepsis.
- To evaluate the performance of PGMs using various combinations of clinical data.
- To assess the feasibility of PGMs as a diagnostic aid in the pediatric intensive care unit (PICU).
Main Methods:
- A retrospective analysis of 3,014 pediatric ICU admissions (2010-2019) was performed.
- Tree Augmented Naive Bayes, a PGM, was employed to construct diagnostic models.
- Models integrated data from vital signs, clinical symptoms, laboratory tests, and microbiological tests, with clinician-selected variables.
Main Results:
- The study included 134 sepsis and 2,880 non-sepsis cases.
- All PGM models demonstrated high accuracy (0.92-0.96) and specificity (0.95-0.99).
- The model incorporating all four data categories achieved the highest performance (accuracy: 0.93, AUC: 0.87), though sensitivity varied (0.46).
Conclusions:
- Probabilistic graphical models are a feasible tool for aiding pediatric sepsis diagnosis.
- The combined model showed strong diagnostic performance, suggesting clinical utility.
- Further validation with diverse datasets is recommended to confirm PGM effectiveness in sepsis diagnosis.
Background:
Probabilistic graphical model, a rich graphical framework in modelling associations between variables in complex domains, can be utilized to aid clinical diagnosis. However, its application in pediatric sepsis remains limited. This study aims to explore the utility of probabilistic graphical models in pediatric sepsis in the pediatric intensive care unit.
Methods:
We conducted a retrospective study on children using the first 24-hour clinical data of the intensive care unit admission from the Pediatric Intensive Care Dataset, 2010-2019. A probabilistic graphical model method, Tree Augmented Naive Bayes, was used to build diagnosis models using combinations of four categories: vital signs, clinical symptoms, laboratory, and microbiological tests. Variables were reviewed and selected by clinicians. Sepsis cases were identified with the discharged diagnosis of sepsis or suspected infection with the systemic inflammatory response syndrome. Performance was measured by the average sensitivity, specificity, accuracy, and area under the curve of ten-fold cross-validations.
Results:
We extracted 3,014 admissions [median age of 1.13 (interquartile range: 0.15-4.30) years old]. There were 134 (4.4%) and 2,880 (95.6%) sepsis and non-sepsis patients, respectively. All diagnosis models had high accuracy (0.92-0.96), specificity (0.95-0.99), and area under the curve (0.77-0.87). Sensitivity varied with different combinations of variables. The model that combined all four categories yielded the best performance [accuracy: 0.93 (95% confidence interval (CI): 0.916-0.936); sensitivity: 0.46 (95% CI: 0.376-0.550), specificity: 0.95 (95% CI: 0.940-0.956), area under the curve: 0.87 (95% CI: 0.826-0.906)]. Microbiological tests had low sensitivity (<0.10) with high incidence of negative results (67.2%).
Conclusions:
We demonstrated that the probabilistic graphical model is a feasible diagnostic tool for pediatric sepsis. Future studies using different datasets should be conducted to assess its utility to aid clinicians in the diagnosis of sepsis.
