Related Experiment Video
Updated: Jun 29, 2026
Separation of Immune Cell Subpopulations in Peripheral Blood Samples from Children with Infectious Mononucleosis
Published on: September 7, 2022
Development and external validation of a machine learning prediction model for Epstein-Barr virus-associated
Li Xiao1, Meiling Liao1, Yan Meng2
1Big Data Center for Children's Medical Care, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, Children's Hospital of Chongqing Medical University, Intelligent Application of Big Data in Pediatrics Engineering Research Center of Chongqing Education Commission of China, Chongqing, China.
Insights
A new machine learning model accurately predicts Epstein-Barr virus-associated hemophagocytic lymphohistiocytosis (EBV-HLH) in children using routine blood tests. This tool aids early diagnosis of life-threatening EBV-HLH versus self-limiting EBV-associated infectious mononucleosis (EBV-IM).
Area of Science:
- Pediatric Infectious Diseases
- Machine Learning in Medicine
- Hematology
Background:
- Epstein-Barr virus (EBV) infection is common in children, presenting challenges in differentiating infectious mononucleosis (IM) from life-threatening hemophagocytic lymphohistiocytosis (HLH).
- Early distinction between EBV-associated IM (EBV-IM) and EBV-associated HLH (EBV-HLH) is critical for timely intervention, yet reliable prediction models are lacking.
- This study addresses the need for accessible diagnostic tools using readily available laboratory parameters.
Purpose of the Study:
- To develop and validate a machine learning (ML) model for predicting EBV-HLH in pediatric patients.
- To utilize routine complete blood count (CBC) parameters obtained within 24 hours of hospital admission for risk assessment.
- To differentiate EBV-associated HLH from EBV-associated infectious mononucleosis in children.
Main Methods:
- Retrospective cohort study of 4,871 pediatric patients with confirmed acute EBV infection.
- Development and internal testing of 13 ML algorithms, with Random Forest (RF) selected for optimal performance.
- External validation using data from a separate hospital campus; SHAP analysis for model interpretability.
Main Results:
- The RF model achieved high accuracy in both internal (AUC=0.993) and external (AUC=0.971) validation cohorts.
- Key predictors identified by SHAP analysis include White Blood Cell count (WBC), Platelet count (PLT), Lactate (LAC), and Hemoglobin (Hb).
- A free online decision-support tool was developed based on the RF model for real-time risk assessment.
Conclusions:
- The RF-based model effectively assesses admission-based risk for pediatric EBV-HLH using routine CBC parameters.
- The model demonstrates excellent generalizability and offers a cost-effective tool for diverse healthcare settings.
- This ML approach facilitates early identification of EBV-HLH, enabling prompt clinical management.
Background:
Epstein-Barr virus (EBV) infection is a common pediatric infectious disease. Infectious mononucleosis (IM) and hemophagocytic lymphohistiocytosis (HLH), two major complications of EBV infection, share similar clinical manifestations in the early stage. While IM is typically self-limiting, HLH is life-threatening and requires immediate intervention. Early differentiation between these two conditions is crucial for clinical decision-making; however, reliable prediction models based on readily available laboratory parameters remain scarce. This study aimed to develop and validate a machine learning prediction model using routine blood parameters obtained within 24 h of hospital admission in children with confirmed acute EBV infection to distinguish EBV-associated IM (EBV-IM) from EBV-associated HLH (EBV-HLH).
Methods:
This retrospective cohort study included 4,871 pediatric patients diagnosed with either EBV-IM or EBV-HLH from two campuses of Children's Hospital of Chongqing Medical University. Demographic information and initial complete blood count (CBC) parameters within 24 h of admission were collected. The cohort was divided into a model development group (Yuzhong Campus, n = 2,848; 70% for training, 30% for internal testing) and an external validation group (Liangjiang Campus, n = 2,023). Thirteen machine learning algorithms were evaluated using random search with 5-fold cross-validation for hyperparameter tuning. Shapley Additive exPlanations (SHAP) analysis was performed to interpret model predictions.
Results:
EBV-HLH accounted for 12.46% (607/4,871) of the total cohort, with significantly different prevalence between the development and validation cohorts (18.29% vs. 4.25%, p < 0.001). Significant differences were observed between cohorts in age and all CBC parameters except gender (p < 0.05). The Random Forest (RF) model demonstrated optimal performance in the internal validation set (AUC = 0.993, 95% CI: 0.990-0.996). In the external validation cohort, the RF model maintained robust discriminative ability (AUC = 0.971, 95% CI: 0.949-0.992). Calibration curves indicated excellent agreement between predicted probabilities and actual risks. SHAP analysis identified WBC, PLT, LAC, and Hb as the most critical predictors of EBV-HLH. DCA demonstrated substantial clinical net benefit. A free online decision-support tool ( https://wangrj1988.shinyapps.io/EBV-HLH-IM/ ) was developed based on the RF model to facilitate real-time risk assessment.
Conclusions:
The RF-based model using routine CBC parameters enables admission-based risk assessment of pediatric EBV-HLH with excellent generalizability, offering a cost-effective tool for diverse healthcare settings.
Trial Registration:
Clinical trial number: not applicable.