Clinical prediction models for pediatric Epstein-Barr virus infectious mononucleosis: from 6 machine learning algorithms Running Title: Machine learning Models for Pediatric EBV Infectious Mononucleosis

Clinical prediction models for pediatric Epstein-Barr virus infectious mononucleosis: from 6 machine learning algorithms Running Title: Machine learning Models for Pediatric EBV Infectious Mononucleosis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Clinical prediction models for pediatric Epstein-Barr virus infectious mononucleosis: from 6 machine learning algorithms Running Title: Machine learning Models for Pediatric EBV Infectious Mononucleosis Ruibing Zhao, Ce Wang, Qian Tian, Nan Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8437947/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Background Pediatric Epstein-Barr virus (EBV) Infectious mononucleosis (PEBV-IM) is an acute infectious disease. However, there are no effective clinical diagnostic indicators for PEBV-IM. The study aimed to construct a clinical model for effective prediction of PEBV-IM and to identify relevant key feature variables, thereby providing a favorable clinical decision-making tool for PEBV-IM. Methods Data were obtained from the clinical diagnosis of PEBV-IM patients, and the feature variables were acquired by least absolute shrinkage and selection operator (LASSO) regression analysis. Subsequently, optimal clinical prediction models and key feature variables for PEBV-IM were acquired using 6 machine learning (ML) algorithms. Finally, to further investigate the relationship between optimal clinical prediction models and key feature variables, the SHapley Additive exPlanations (SHAP) model interpretation was proceeded. Results A total of 60 PEBV-IM samples and 41 variables were included in the analyses, and 12 characteristic feature variables were identified by LASSO. Subsequently, founded on the feature variables, the clinical prediction model was constructed using the plsRglm algorithm, which achieved the highest accuracy in both the training set (area under the curve (AUC) = 0.939) and the validation set (AUC = 0.850). Thus, this model was identified as the optimal clinical prediction model, while key feature variables platelet count and gamma-glutamyl transferase (GGT) were acquired. Notably, the GGT had the significant effect on the output of the clinical prediction model, with low GGT having a positive effect on the output, while low feature values of platelet count had a negative effect on the model output. Conclusion Obtaining a highly accurate clinical prediction model for PEBV-IM and 2 key feature variables (platelet count and GGT), which, in combination with SHAP model interpretation, provided a clear understanding and a novel tool for early diagnosis and clinical decision-making in PEBV-IM. Epstein-Barr virus Infectious mononucleosis Pediatric Epstein-Barr virus infectious mononucleosis Clinical prediction model Machine learning SHAP Figures Figure 1 Figure 2 Figure 3 1. Introduction Epstein-Barr virus (EBV), also known as human herpesvirus 4(HHV-4), is a member of the gamma herpesvirus family that infects over 95% of the world's population 1 . EBV was first identified in Burkitt lymphoma (BL) in 1964 and was later found to be associated with other types of lymphoma, including Hodgkin lymphoma (HL), non-HL, T cell lymphoma, and natural killer (NK)/T cell lymphoma in posttransplant patients and HIV infected individuals. EBV is also linked to epithelial cancers, such as nasopharyngeal carcinoma (NPC) and gastric cancers. Additionally, EBV is associated with nonmalignant diseases, including infectious mononucleosis (IM), oral hairy leukoplakia, systemic lupus erythematosus (SLE), and multiple sclerosis (MS) 2 . Although several preventive or therapeutic vaccine strategies for EBV have been evaluated in clinical and preclinical trials, no licensed vaccines or therapeutic interventions for EBV-associated diseases 3–5 . Primary EBV infection is often asymptomatic, but when the immune response is strong, the resulting disease state is called IM. Patients with symptomatic IM typically experience the classic triad of fever, lymphadenopathy, and pharyngitis 6,7 , along with other symptoms such as transient hepatitis, splenomegaly, malaise, nausea, and palatal petechiae, depending on the host's response to the invading virus 8 . Painless bilateral swelling of the upper eyelids 9,10 and thrombocytopenia 11–13 may also occur. Although IM is a self-limiting disease and symptoms usually resolve within weeks, fatigue may persist for months. Early, accurate laboratory results are essential for the correct diagnosis of IM, allowing for timely intervention and avoiding unnecessary treatments, such as antibiotics for pharyngitis symptoms or expensive exploratory tests for cases with splenomegaly or suspicious hematological disease 14 . Therefore, reliable diagnostic metrics and clinical predictive models are crucial for effective decision-making in PEBV-IM. Machine Learning (ML) enables computer systems to learn from data, improve performance, and make predictions or decisions by identifying patterns and building mathematical models. Common methods used in clinical prediction models include linear regression, logistic regression, support vector machines, decision trees, random forests, neural networks. These methods can process large-scale and complex data, improve prediction accuracy, personalize treatment, and provide real-time decision-making tools for PEBV-IM. Based on clinical and pathological data collected from PEBV-IM patients, this study constructs a clinical predictive model for PEBV-IM using bioinformatics analysis and multiple ML algorithms, identifying key variables related to PEBV-IM. This model provides a valuable decision-making tool for PEBV-IM, enabling early intervention and personalized treatment. 2. Materials and methods 2.1 Data collection and processing Data on clinical characteristics of pediatric Epstein-Barr virus (EBV) infectious mononucleosis (PEBV-IM) patients were collected from Xi'an Children's Hospital, and informed consent was obtained from the patient’s guardian for the used of the data. Firstly, the PEBV-IM dataset contained information on totally 99 cases with the disease, subsequently, variables with null data (days to EBV diagnosis, cluster of differentiation 25 (CD25), natural killer (NK) cell activity), variables with only one status (lymph node enlargement, history of immunosuppression or not), and variables with too large a disparity in sample size (sore throat / pharyngeal isthmus (yes: 97, no: 2), presence of haemophilia on bone puncture (not investigated: 95, yes: 2, no : 2)) were excluded. Finally, the 39 samples in which EBV load were not checked were excluded, a sum of 60 samples and 41 variables included in the analysis ( Supplementary Table 1 ). 2.2 Baseline characteristics analysis The PEBV-IM dataset was first split into training and validation sets in a 7: 3 ratio by random splitting. Subsequently, so as to assess whether there were marked differences in the clinical feature variables between the training and validation sets. Subsequently, the quartile method 15 was adopted to split continuous variables into categorical variables (Q1-Q4), and the Mann-Whitney U test was applied to compare differences in clinical characteristics. Then, the chi-square test was applied to compare the differences in clinical characteristics of categorical variables, excluding variables with marked in the both dataset (p < 0.05). 2.3 Construction of clinical prediction models To identify feature variables for predicting PEBV-IM, founded on training set, the aspartate aminotransferase (AST) / alanine aminotransferase (ALT) was designated as the dependent variable (to be 0 for less than the median and 1 for greater than the median). Then, the glmnet package (v 4.1-8) 16 was adopted to proceed 10-fold cross-validation least absolute shrinkage and selection operator (LASSO) regression analysis. The variables with the smallest model error and the regression coefficient not penalized to 0 were selected as the feature variables. Moreover, founded on the feature variables, the NaiveBayes, RF, SVM, Ridge, plsRglm and LDA algorithms from e1071 package (v 1.7–14), 17 randomForestSRC package (v 3.3.0) 18 , e1071 package (v 1.7–14), glmnet package (v 4.1-8), plsRglm package (v 1.5.1) 19 , and MASS package (v 7.3–60.0.1 https://cran.r-project.org/web/packages/MASS/index.html ) were applied to construct the clinical predictive models, respectively. Subsequently, in the training and validation sets, the pROC package (v 1.18.0) 20 was applied to evaluate the accuracy of predictive models by plotting receiver operating characteristic (ROC) curves, and the optimal clinical predictive model of PEBV-IM with the highest the area under curve (AUC) values in both datasets was selected. Further, founded on the optimal clinical prediction model, the key feature variables were acquired. 2.4 SHapley Additive exPlanations (SHAP) model interpretation So as to better understand the relationship between the key feature variables and the clinical prediction model, the shapviz package (v 0.9.5) 21 was applied to compute the mean |SHAP| values of the key feature variables and to acquire distribution of feature importance. Subsequently, the key feature variables were ranked in descending order of importance founded on the mean |SHAP| values of the key feature variables, the ggplot2 package (v 3.4.2) 22 was applied to visualize the results. Moreover, the impact of single key feature variables on the optimal clinical prediction model results was observed. 2.5 Statistical analysis All statistical analyses were conducted using R software (v 4.3.1). The Mann-Whitney U test and chi-square test were utilized to evaluate paired sample comparisons, with statistical significance defined as p < 0.05. 3. Results 3.1 The 39 variables were applied in subsequent analyses A sum of 60 PEBV-IM samples and 41 variables were included in the study, with the training set and validation set containing 47 and 13 PEBV-IM samples, respectively. Then, the results of the baseline characteristics analysis displayed that the T cells and Ts cells count were markedly different in the training and validation sets (p 0.05) and could be applied for subsequent analyses (Table 1 ). 3.2 The optimal clinical predictive models and key characteristic variables were obtained In the training set, founded on the 39 variables, 12 feature variables were acquired by LASSO regression analysis (log(lambda.min) = -2.9235 (Fig. 1 A, B), namely age, gender, degree of splenomegaly, percentage of lymphocytes, red blood cells count, hemoglobin, platelet count, GGT, helper T (Th) cells count, B cells count, interleukin (IL)-1 beta (IL-1β) and IL-10 (Table 2 ). Subsequently, 6 clinical prediction models were acquired founded on the construction of 12 feature variables (Table 3 ), and it could be observed that the clinical prediction model constructed by plsRglm had the highest AUC values in the training (AUC = 0.939) and validation sets (AUC = 0.850) (Fig. 2 A-C), which was the optimal clinical predictive model for PEBV-IM. Furthermore, platelet count and GGT were identified as key feature variables. Table 1 Baseline characterisation Variable Level Validation set Training set p n 13 47 Age (%) Q1 2 (15.4) 13 (27.7) 0.449 Q2 2 (15.4) 13 (27.7) Q3 5 (38.5) 10 (21.3) Q4 4 (30.8) 11 (23.4) Gender (%) man 7 (53.8) 25 (53.2) 1.000 woman 6 (46.2) 22 (46.8) Highest body temperature (%) Q1 2 (15.4) 13 (27.7) 0.717 Q2 1 ( 7.7) 6 (12.8) Q3 6 (46.2) 16 (34.0) Q4 4 (30.8) 12 (25.5) Days with body temperature > 38℃ (%) Q1 0 ( 0.0) 8 (17.0) 0.318 Q2 3 (23.1) 6 (12.8) Q3 7 (53.8) 19 (40.4) Q4 3 (23.1) 14 (29.8) Degree of liver enlargement (%) Light 6 (46.2) 20 (42.6) 0.964 Moderate 2 (15.4) 7 (14.9) Normal 5 (38.5) 20 (42.6) Degree of splenomegaly (%) Light 2 (15.4) 6 (12.8) 0.666 Moderate 7 (53.8) 20 (42.6) Normal 4 (30.8) 21 (44.7) Epstein-Barr virus (EBV) load (%) < 5×10 2 6 (46.2) 21 (44.7) 0.513 5×10 3 ~5×10 4 3 (23.1) 16 (34.0) 5×10 4 ~5×10 5 2 (15.4) 2 ( 4.3) 5×10 2 ~5×10 2 2 (15.4) 8 (17.0) White blood cells count (%) Q1 2 (15.4) 13 (27.7) 0.601 Q2 3 (23.1) 12 (25.5) Q3 5 (38.5) 10 (21.3) Q4 3 (23.1) 12 (25.5) Neutrophil count (%) Q1 2 (15.4) 13 (27.7) 0.782 Q2 4 (30.8) 11 (23.4) Q3 4 (30.8) 11 (23.4) Q4 3 (23.1) 12 (25.5) Lymphocyte count (%) Q1 1 ( 7.7) 14 (29.8) 0.300 Q2 5 (38.5) 10 (21.3) Q3 4 (30.8) 10 (21.3) Q4 3 (23.1) 13 (27.7) Percentage of lymphocytes (%) Q1 3 (23.1) 12 (25.5) 0.961 Q2 3 (23.1) 12 (25.5) Q3 4 (30.8) 11 (23.4) Q4 3 (23.1) 12 (25.5) Continued Table 1 Baseline characterisation Variable Level Validation set Training set p Red blood cells count (%) Q1 4 (30.8) 11 (23.4) 0.961 Q2 3 (23.1) 12 (25.5) Q3 3 (23.1) 12 (25.5) Q4 3 (23.1) 12 (25.5) Hemoglobin (%) Q1 3 (23.1) 10 (21.3) 0.637 Q2 3 (23.1) 12 (25.5) Q3 2 (15.4) 14 (29.8) Q4 5 (38.5) 11 (23.4) Platelet count (%) Q1 5 (38.5) 10 (21.3) 0.601 Q2 3 (23.1) 12 (25.5) Q3 3 (23.1) 12 (25.5) Q4 2 (15.4) 13 (27.7) Triglycerides (%) Q1 3 (23.1) 11 (23.4) 0.348 Q2 4 (30.8) 12 (25.5) Q3 5 (38.5) 10 (21.3) Q4 1 ( 7.7) 14 (29.8) Fibrinogen (%) Q1 4 (30.8) 9 (19.1) 0.839 Q2 3 (23.1) 14 (29.8) Q3 3 (23.1) 12 (25.5) Q4 3 (23.1) 12 (25.5) Total bilirubin (%) Q1 2 (15.4) 13 (27.7) 0.449 Q2 2 (15.4) 13 (27.7) Q3 4 (30.8) 11 (23.4) Q4 5 (38.5) 10 (21.3) Direct bilirubin (%) Q1 2 (15.4) 13 (27.7) 0.122 Q2 4 (30.8) 11 (23.4) Q3 1 ( 7.7) 14 (29.8) Q4 6 (46.2) 9 (19.1) Alanine aminotransferase (ALT) (%) Q1 2 (15.4) 13 (27.7) 0.782 Q2 4 (30.8) 11 (23.4) Q3 4 (30.8) 11 (23.4) Q4 3 (23.1) 12 (25.5) Aspartate aminotransferase (AST) (%) Q1 4 (30.8) 11 (23.4) 0.747 Q2 2 (15.4) 12 (25.5) Q3 4 (30.8) 10 (21.3) Q4 3 (23.1) 14 (29.8) Lactate dehydrogenase (LDH) (%) Q1 3 (23.1) 12 (25.5) 0.329 Q2 5 (38.5) 10 (21.3) Q3 1 ( 7.7) 14 (29.8) Q4 4 (30.8) 11 (23.4) Continued Table 1 Baseline characterisation Variable Level Validation set Training set p Gamma-glutamyl transferase (GGT) (%) Q1 2 (15.4) 12 (25.5) 0.446 Q2 2 (15.4) 14 (29.8) Q3 5 (38.5) 10 (21.3) Q4 4 (30.8) 11 (23.4) Albumin (%) Q1 3 (23.1) 12 (25.5) 0.601 Q2 3 (23.1) 12 (25.5) Q3 5 (38.5) 10 (21.3) Q4 2 (15.4) 13 (27.7) Ferritin (%) Q1 3 (23.1) 11 (23.4) 0.681 Q2 5 (38.5) 11 (23.4) Q3 3 (23.1) 12 (25.5) Q4 2 (15.4) 13 (27.7) T cells count (%) Q1 1 ( 7.7) 14 (29.8) 0.021 Q2 1 ( 7.7) 14 (29.8) Q3 4 (30.8) 11 (23.4) Q4 7 (53.8) 8 (17.0) Suppressor T (Ts) cells count (%) Q1 1 ( 7.7) 14 (29.8) 0.043 Q2 1 ( 7.7) 14 (29.8) Q3 5 (38.5) 10 (21.3) Q4 6 (46.2) 9 (19.1) Helper T (Th) cells count (%) Q1 3 (23.1) 12 (25.5) 0.601 Q2 5 (38.5) 10 (21.3) Q3 3 (23.1) 12 (25.5) Q4 2 (15.4) 13 (27.7) Natural killer (NK) cells count (%) Q1 3 (23.1) 12 (25.5) 0.782 Q2 2 (15.4) 13 (27.7) Q3 4 (30.8) 11 (23.4) Q4 4 (30.8) 11 (23.4) B cells count (%) Q1 5 (38.5) 10 (21.3) 0.329 Q2 4 (30.8) 11 (23.4) Q3 3 (23.1) 12 (25.5) Q4 1 ( 7.7) 14 (29.8) Interleukin (IL) -1 beta (IL-1β) (%) Q1 4 (30.8) 10 (21.3) 0.259 Q2 1 ( 7.7) 13 (27.7) Q3 5 (38.5) 9 (19.1) Q4 3 (23.1) 15 (31.9) IL-2 (%) Q1 0 ( 0.0) 8 (17.0) 0.327 Q2 5 (38.5) 17 (36.2) Q3 5 (38.5) 10 (21.3) Q4 3 (23.1) 12 (25.5) Continued Table 1 Baseline characterisation Variables Level Validation set Training set p IL-4 (%) Q1 2 (15.4) 13 (27.7) 0.616 Q2 1 ( 7.7) 2 ( 4.3) Q4 10 (76.9) 32 (68.1) IL-5 (%) Q1 2 (15.4) 5 (10.6) 0.637 Q3 9 (69.2) 29 (61.7) Q4 2 (15.4) 13 (27.7) IL-6 (%) Q1 2 (15.4) 13 (27.7) 0.727 Q2 4 (30.8) 10 (21.3) Q3 3 (23.1) 13 (27.7) Q4 4 (30.8) 11 (23.4) IL-8 (%) Q1 1 ( 7.7) 4 ( 8.5) 0.655 Q2 7 (53.8) 17 (36.2) Q3 3 (23.1) 12 (25.5) Q4 2 (15.4) 14 (29.8) IL-10 (%) Q1 3 (23.1) 11 (23.4) 0.816 Q2 4 (30.8) 12 (25.5) Q3 2 (15.4) 13 (27.7) Q4 4 (30.8) 11 (23.4) Interleukin-12 p70 subunit (IL-12p70) (%) Q1 2 (15.4) 5 (10.6) 0.263 Q3 10 (76.9) 28 (59.6) Q4 1 ( 7.7) 14 (29.8) IL-17 (%) Q1 2 (15.4) 10 (21.3) 0.304 Q2 2 (15.4) 5 (10.6) Q3 8 (61.5) 18 (38.3) Q4 1 ( 7.7) 14 (29.8) Tumor necrosis factor-alpha (TNF-α) (%) Q1 4 (30.8) 8 (17.0) 0.454 Q3 7 (53.8) 26 (55.3) Q4 2 (15.4) 13 (27.7) Interferon-gamma (IFN-γ) (%) Q1 4 (30.8) 11 (23.4) 0.782 Q2 2 (15.4) 13 (27.7) Q3 4 (30.8) 11 (23.4) Q4 3 (23.1) 12 (25.5) Interferon-alpha (IFN-α )(%) Q1 4 (30.8) 11 (23.4) 0.145 Q2 0 ( 0.0) 14 (29.8) Q3 5 (38.5) 10 (21.3) Q4 4 (30.8) 12 (25.5) Table 2 Characteristic Variable Analysis of PEBV-IM Feature variable Coefficient Age -0.02434 Gender -0.38919 Degree of splenomegaly -0.14292 Percentage of lymphocytes 0.01561 Red blood cells count 0.50542 hemoglobin 0.00313 Platelet count 0.00491 GGT -0.01484 Th cells count 0.00005 B cells count 0.00005 IL-1β -0.00874 IL-10 -0.02497 Table 3 Machine learning to acquire key feature variables GGT RF Platelet count RF hemoglobin RF Red blood cells count RF B cells count RF Age RF Age NaiveBayes Gender NaiveBayes Degree of splenomegaly NaiveBayes Percentage of lymphocytes NaiveBayes Red blood cells count NaiveBayes hemoglobin NaiveBayes Platelet count NaiveBayes GGT NaiveBayes Th cells count NaiveBayes B cells count NaiveBayes IL-1β NaiveBayes IL-10 NaiveBayes Age SVM Gender SVM Degree of splenomegaly SVM Percentage of lymphocytes SVM Red blood cells count SVM hemoglobin SVM Platelet count SVM GGT SVM Th cells count SVM B cells count SVM IL-1β SVM IL-10 SVM Age Ridge Gender Ridge Degree of splenomegaly Ridge Percentage of lymphocytes Ridge Continued Table 3 Machine learning to acquire key feature variables Red blood cells count Ridge hemoglobin Ridge Platelet count Ridge GGT Ridge Th cells count Ridge B cells count Ridge IL-1β Ridge IL-10 Ridge Platelet count plsRglm GGT plsRglm Age LDA Gender LDA Degree of splenomegaly LDA Percentage of lymphocytes LDA Red blood cells count LDA hemoglobin LDA Platelet count LDA GGT LDA Th cells count LDA B cells count LDA IL-1β LDA IL-10 LDA 3.3 GGT and platelet count affected model output The SHAP values could determine the difference between the predicted values for all combinations, by calculating the mean |SHAP| values, it could be observed that the GGT had the greatest effect on the clinical prediction model output (Fig. 3 A). In addition, the effect of features on the model output was explored, the results displayed that samples with high GGT feature values had a negative effect on the prediction, and samples with low GGT feature values had a positive effect on the prediction. On the feature values of platelet count, a low feature values had a negative effect on the model output and conversely a positive effect (Fig. 3 B). Finally, by focusing on the dependence between platelet count and GGT, it was found that both had a linear relationship with the model’s predictions. This indicated that the predictions were not only affected by the level of each key feature variable but also by the individual differences between them (Fig. 3 C). Overall, the interpretability and transparency of the clinical prediction model was improved by SHAP interpretation, which enabled used to better understand the predictive results of the model for PEBV-IM risk prediction. 4. Discussion EBV infection is a widespread condition that affects nearly all tissues and organs in the body 23 , 24 , with systemic effects on the liver, blood and immune system in EBV-associated diseases 25 – 27 . Despite its prevalence, there is a critical lack of reliable predictive models for early identification and intervention in IM, particularly in pediatric populations. The present study addresses this gap by constructing a clinical predictive model using 6 ML algorithms, providing a powerful tool for early diagnosis and intervention in PEBV-IM. Most primary EBV infections manifest as IM in the pediatric population, and in a few cases, they progress to more severe conditions like EBV-associated hemophagocytic lymphohistiocytosis (HLH) 28 . Huang R et al. 29 used LASSO regression and logistic regression analysis to identify risk factors and independent predictors for EBV, while also constructing nomograms to assist in early detection. In addition, a small minority of individuals may develop chronically active Epstein-Barr virus (CAEBV) infection, after EBV infection, leading to a disease prone to relapse and associated with severe and potentially fatal comorbidities such as malignant lymphoma, HLH, multiple organ failure, and disseminated intravascular coagulation (DIC) 30 , 31 . Yuan JH et al. 32 enhanced their results by applying LASSO Logistic regression to identify key peripheral blood biomarkers that distinguish IM from CAEBV, thus offering valuable guidance for clinical work. Both studies highlight the reliability of LASSO as a high-dimensional predictor regression method 33 . Similarly, in this study, 39 clinical variables were initially analyzed, with 12 characteristic variables identified through LASSO regression analysis. This led to the construction of 6 clinical predictive models, ultimately identifying GGT and platelet count as key characteristic variables for predicting PEBV-IM. GGT is a glycosylated protein located on the outer plasma membrane, a microsomal enzyme involved in the homeostasis of glutathione and cysteine 34 . GGT plays a crucial role in transferring glutamyl from gamma-glutamate to other peptides, and although its activity in the kidney is ten times higher than in the liver, serum GGT is typically used as a liver test 35 . Studies have shown that about 85% of EBV-infected patients experience varying degrees of liver impairment, with some cases leading to hepatitis, which increases transaminase levels and impairs liver function 36 . HeHC et al. 37 observed consistently elevated GGT levels in EBV-infected gastric cancer patients, while ZhangL 38 found GGT to be significantly higher in IM patients than in healthy controls, demonstrating its diagnostic and predictive value. These findings are consistent with our results, where high GGT characteristics were negatively associated with the prediction of IM, indicating that patients with higher GGT levels are more likely to be diagnosed with IM and experience longer hospital stays. Monitoring GGT can thus help help identify patients at greater risk and allow for early intervention. Platelet are essential for coagulation and the repair of vascular endothelial cells, with significant decreases in platelet count leading to spontaneous bleeding, which can be life-threatening in some severe cases 39 . Previous studies have suggested that IM can cause thrombocytopenia in children 40 . A recent study reported by Zhang et al described a 14-year-old girl who developed severe thrombocytopenia (with a platelet count down to 5×10 9 /L), leading to spontaneous bleeding and periorbital edema, a rare symptom of IM 41 . Yan et al. reported EBV-DNA positivity in blood lymphocytes of patients with chronic immune thrombocytopenia (detected by quantitative reverse transcription PCR) higher than controls 42 . LiY et al. 43 found in a study that EBV-DNA copy number in plasma can be used as an important indicator to assess the duration of fever and clinical outcomes in children. These researches showed that platelet counts correlate with EBV-DNA copy number in plasma, suggesting that this indicator can be used to assess the duration of fever and clinical outcomes in children. HanX et al. 44 used regression analysis to analyze the predictive value of platelet four indicators in IM patients, and established three regression models for regression analysis, showing that PLT is a significant predictor of IM. LinJ et al. 45 proposed in their study that L and platelet count < 50 × 10 12 /L are independent risk factors for mortality in CAEBV patients. PrtorićL et al. 46 reported in a recent article that each unit decrease in platelet count increases hospital stay by 0.12 days, so platelet count is a predictor of hospital stay. Our study supports these findings, with low platelet eigenvalues negatively impacting the model output, and vice versa. That is, the lower the platelet number, the more severe the child's condition and the longer the hospital stay. Therefore, platelet count can be used to clinically assess the severity of IM in children, helping clinicians identify those at risk and intervene early. In this study, we developed the best clinical predictive model for PEBV-IM through 6 ML methods, particularly the plsRglm model. However, a major challenge in the medical application of ML models is the lack of interpretability 47 . To address this, we introduced SHAP values, which were proposed by Lundberg et al. 48 for accurately calculating the contribution and impact (positive or negative) of each feature on the final prediction. SHAP can perform both local and global interpretability and has a solid theoretical foundation compared to other methods 49 . In this study, SHAP values allowed us to better understand the impact of key characteristic variables (GGT and platelet count) on the model output. Interestingly, the relationship between GGT and platelet count was linear, with higher GGT characteristic value and lower platelet characteristic value being associated with more severe IM. This further emphasizes that predictions were influenced not only by the level of each key characteristic variable but also by their individual differences between them. ML, especially in healthcare, holds great promise in handling large, complex datasets to improve predictions. Greener JG 50 highlighted the importance of matching predictive models to data and automating analysis to build repeatable and time-saving pipelines. ML algorithms are particularly useful when dealing with vast amounts of clinical data, as demonstrated by their widespread application in discovering diagnostic markers and predicting disease outcomes 51 . Our study highlights the potential of ML in clinical settings, particularly for identifying key biomarkers and predicting disease progression in IM patients. This could ultimately lead to more personalized and effective treatments. Despite the promising results, there are several limitations in this study that warrant consideration. First, the sample size is relatively small, which may limit the generalizability and robustness of the model. Larger and multi-center studies are needed to validate the findings and enhance the model's applicability across diverse clinical settings. Second, some variables were missing in the dataset, which could introduce bias into the analysis. Lastly, while ML models are powerful tools, their complexity and the difficulty in interpreting results may hinder their repeatability and widespread use in clinical practice. Future research should focus on improving the model’s interpretability, expanding the clinical data set, and conducting clinical trials to verify its real-world applicability. In conclusion, our study demonstrates the potential of ML in creating an effective predictive model for PEBV-IM. The identification of key characteristic variables (GGT and platelet count) provides valuable insights for early diagnosis and intervention. By integrating these predictive tools into clinical practice, we can improve patient outcomes through more personalized and timely interventions. Future studies should aim to refine and validate these models for broader clinical use. Abbreviations abbreviation full name EBV Epstein-Barr virus PEBV-IM Pediatric Epstein-Barr virus Infectious mononucleosis LASSO least absolute shrinkage and selection operator ML machine learning SHAP SHapley Additive exPlanations AUC area under the curve GGT gamma-glutamyl transferase HHV-4 human herpes virus 4 BL Burkitt lymphoma HL Hodgkin lymphoma NK natural killer NPC nasopharyngeal carcinoma IM infectious mononucleosis SLE systemic lupus erythematosus MS multiple sclerosis ML Machine Learning CD25 cluster of differentiation 25 AST aspartate aminotransferase ALT alanine aminotransferase ROC receiver operating characteristic Th helper T LDH Lactate dehydrogenase Ts Suppressor T IL Interleukin IL-1β Interleukin -1 beta IL-12p70 Interleukin-12 p70 subunit TNF-α Tumor necrosis factor-alpha IFN-γ Interferon-gamma IFN-α Interferon-alpha HLH hemophagocytic lymphohistiocytosis CAEBV chronically active Epstein-Barr virus DIC disseminated intravascular coagulation Declarations Ethics approval and consent to participate The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethical Committee for Drug Clinical Trials at Xi 'an Children's Hospital (Ethics Approval No.: 20260036). Informed consent has been obtained from the patient.For the minors under 16 years old in this study, written informed consent has been obtained from their parents (or legal guardians). Consent for publication Not applicable Competing interests The authors declare that they have no competing interests Clinical trial number Not applicable Funding Not applicable Author Contribution CeWang:Writing-review& editing, Conceptualization, Methodology, Project administration;RuibingZhao: Writing – review & editing, Conceptualization, Methodology; CeWang: Writing – original draft, Conceptualization, Methodology; RuibingZhao: Writing – review & editing, Formal Analysis; QianTian: Writing – review & editing; Nan Wang: Writing – review & editing Acknowledgements We would like to express our sincere gratitude to all individuals and organizations who supported and assisted us throughout this research. Special thanks to the following authors: In conclusion, we extend our thanks to everyone who has supported and assisted us along the way. Without your support, this research would not have been possible. Data Availability The datasets generated and/or analyzed during the current study are not publicly available due to the protection of patients' privacy, but can be obtained from the corresponding author upon reasonable request. References Damania B, Kenney SC, Raab-Traub N. Epstein-Barr virus: Biology and clinical disease. Cell. 2022;185(20):3652–70. 10.1016/j.cell.2022.08.026 . Gewurz BL, Longnecker RM, Cohen JI. (2021). Epstein-Barr virus(Chap. 11). Fields Virology, 324–388. Smith C, Khanna R. The development of prophylactic and therapeutic EBV vaccines. Curr Top Microbiol Immunol. 2015;391:455–73. 10.1007/978-3-319-22834-1_16 . Cohen JI. Vaccine development for Epstein-Barr virus. Adv Exp Med Biol. 2018;1045:477–93. 10.1007/978-981-10-7230-7_22 . Dasari V, Bhatt KH, Smith C, Khanna R. Designing an effective vaccine to prevent Epstein-Barr virus-associated diseases: challenges and opportunities. Expert Rev Vaccines. 2017;16(4):377–90. 10.1080/14760584.2017.1293529 . HOAGLAND RJ. The clinical manifestations of infectious mononucleosis: a report of two hundred cases. Am J Med Sci. 1960;240:55–63. Ming Y, Cheng S, Chen Z, et al. Infectious mononucleosis in children and differences in biomarker levels and other features between disease caused by Epstein-Barr virus and other pathogens: a single-center retrospective study in China. PeerJ. 2023;11:e15071. 10.7717/peerj.15071 . Published 2023 Apr 6. Gao Y, Li L, Hu X, Zhang W, Li Y. Interleukin-35 has a protective role in infectious mononucleosis-induced liver inflammation probably by inhibiting CD8 + T cell function. Arch Immunol Ther Exp (Warsz). 2022;70(1):25. Published 2022 Oct 11. 10.1007/s00005-022-00663-8 Ricardo D. A Protracted Course of Periorbital Oedema in Infectious Mononucleosis Caused by Epstein-Barr Virus. Infect Dis Rep. 2022;14(6):942–5. 10.3390/idr14060092 . Published 2022 Nov 23. Bronz G, Zanetti BPESM, Bianchetti MG, et al. Bilateral upper eyelid swelling (Hoagland sign) in Epstein-Barr infectious mononucleosis: prospective experience. Infection. 2023;51(2):471–4. 10.1007/s15010-022-01932-6 . Páez-Guillán EM, Campos-Franco J, Alende R, Gonzalez-Quintela A. Hematological abnormalities beyond lymphocytosis during infectious mononucleosis: Epstein-Barr virus-induced thrombocytopenia. Mediterr J Hematol Infect Dis. 2023;15(1):e2023023. Published 2023 Mar 1. 10.4084/MJHID.2023.023 Zhang C, Kelly AM. Severe Thrombocytopenia in a case of Epstein-Barr virus-induced infectious mononucleosis. Cureus. 2021;13(9):e17706. 10.7759/cureus.17706 . Published 2021 Sep 4. Yusuf H, Kou A, Zelinskas C, et al. Secondary immune thrombocytopenic purpura due to primary Epstein- Barr virus infection. Cureus. 2022;14(6):e26112. 10.7759/cureus.26112 . Published 2022 Jun 20. Naughton P, Healy M, Enright F, Lucey B. Infectious Mononucleosis: diagnosis and clinical interpretation. Br J Biomed Sci. 2021;78(3):107–16. 10.1080/09674845.2021.1903683 . Ruilope LM, Agarwal R, Anker SD, et al. Blood Pressure and Cardiorenal Outcomes With Finerenone in Chronic Kidney Disease in Type 2 Diabetes. Hypertension. 2022;79(12):2685–95. 10.1161/HYPERTENSIONAHA.122.19744 . Friedman J, Hastie T, Tibshirani R. Regularization Paths for Generalized Linear Models via Coordinate Descent. J Stat Softw. 2010;33(1):1–22. Cinelli M, Sun Y, Best K, et al. Feature selection using a one dimensional naïve Bayes' classifier increases the accuracy of support vector machine classification of CDR3 repertoires. Bioinformatics. 2017;33(7):951–5. 10.1093/bioinformatics/btw771 . Zhao P, Zhen H, Zhao H, Huang Y, Cao B. Identification of hub genes and potential molecular mechanisms related to radiotherapy sensitivity in rectal cancer based on multiple datasets. J Transl Med. 2023;21(1):176. Published 2023 Mar 6. 10.1186/s12967-023-04029-2 Yin Y, Chen C, Zhang D et al. Construction of predictive model of interstitial fibrosis and tubular atrophy after kidney transplantation with machine learning algorithms. Front Genet. 2023;14:1276963. Published 2023 Nov 1. 10.3389/fgene.2023.1276963 Robin X, Turck N, Hainard A, et al. pROC: an open-source package for R and S + to analyze and compare ROC curves. BMC Bioinformatics. 2011;12:77. 10.1186/1471-2105-12-77 . Published 2011 Mar 17. Huang TF, Luo C, Guo LB, et al. Preoperative prediction of textbook outcome in intrahepatic cholangiocarcinoma by interpretable machine learning: A multicenter cohort study. World J Gastroenterol. 2025;31(11):100911. 10.3748/wjg.v31.i11.100911 . Gustavsson EK, Zhang D, Reynolds RH, Garcia-Ruiz S, Ryten M. ggtranscript: an R package for the visualization and interpretation of transcript isoforms using ggplot2. Bioinformatics. 2022;38(15):3844–6. 10.1093/bioinformatics/btac409 . Houen G, Trier NH. Epstein-Barr Virus and systemic autoimmune diseases. Front Immunol. 2021;11:587380. 10.3389/fimmu.2020.587380 . Published 2021 Jan 7. Fugl A, Andersen CL. Epstein-Barr virus and its association with disease - a review of relevance to general practice. BMC Fam Pract. 2019;20(1):62. Published 2019 May 14. 10.1186/s12875-019-0954-3 Zhang C, Cui S, Mao G, Li G. Clinical characteristics and the risk factors of hepatic injury in 221 children with infectious mononucleosis. Front Pediatr. 2022;9:809005. 10.3389/fped.2021.809005 . Published 2022 Jan 12. Vine LJ, Shepherd K, Hunter JG, et al. Characteristics of Epstein-Barr virus hepatitis among patients with jaundice or acute hepatitis. Aliment Pharmacol Ther. 2012;36(1):16–21. 10.1111/j.1365-2036.2012.05122.x . Yang SI, Geong JH, Kim JY. Clinical characteristics of primary Epstein Barr virus hepatitis with elevation of alkaline phosphatase and γ-glutamyltransferase in children. Yonsei Med J. 2014;55(1):107–12. 10.3349/ymj.2014.55.1.107 . Taylor GS, Long HM, Brooks JM, Rickinson AB, Hislop AD. The immunology of Epstein-Barr virus-induced disease. Annu Rev Immunol. 2015;33:787–821. 10.1146/annurev-immunol-032414-112326 . Huang R, Wu D, Wang L, et al. A predictive model for Epstein-Barr virus-associated hemophagocytic lymphohistiocytosis. Front Immunol. 2024;15:1503118. 10.3389/fimmu.2024.1503118 . Published 2024 Dec 5. Naughton P, Healy M, Enright F, Lucey B. Infectious Mononucleosis: diagnosis and clinical interpretation. Br J Biomed Sci. 2021;78(3):107–16. 10.1080/09674845.2021.1903683 . Kim WY, Montes-Mojarro IA, Fend F, Quintanilla-Martinez L. Epstein-Barr virus-associated T and NK-cell lymphoproliferative diseases. Front Pediatr. 2019;7:71. Published 2019 Mar 15. 10.3389/fped.2019.00071 Yuan JH, Pang CJ, Yuan SL. Constructing a predictive model based on peripheral blood signs to differentiate infectious mononucleosis from chronic active EBV infection. J Infect Dev Ctries. 2024;18(9):1429–34. 10.3855/jidc.19233 . Published 2024 Sep 30. Tang G, Qi L, Sun Z, et al. Evaluation and analysis of incidence and risk factors of lower extremity venous thrombosis after urologic surgeries: A prospective two-center cohort study using LASSO-logistic regression. Int J Surg. 2021;89:105948. 10.1016/j.ijsu.2021.105948 . Jiang S, Jiang D, Tao Y. Role of gamma-glutamyltransferase in cardiovascular diseases. Exp Clin Cardiol. 2013;18(1):53–6. Sotil EU, Jensen DM. Serum enzymes associated with cholestasis. Clin Liver Dis. 2004;8(1):41–vi. 10.1016/S1089-3261(03)00136-3 . Mecadon K, Jandovitz N, Salerno D, Martinez M, Kato T, Sammons C. Treatment of Epstein–Barr virus viremia in pediatric intestinal and liver transplant recipients. Transplantation. 2017;101:S135. He HC, Han R, Xu BH et al. Circulating Epstein-Barr virus DNA associated with hepatic impairment and its diagnostic and prognostic role in patients with gastric cancer. Front Med (Lausanne). 2022;9:1028033. Published 2022 Oct 6. 10.3389/fmed.2022.1028033 Zhang L, Zhou P, Meng Z, et al. Infectious mononucleosis and hepatic function. Exp Ther Med. 2018;15(3):2901–9. 10.3892/etm.2018.5736 . Rosado FG, Kim AS. Hemophagocytic lymphohistiocytosis: an update on diagnosis and pathogenesis. Am J Clin Pathol. 2013;139(6):713–27. 10.1309/AJCP4ZDKJ4ICOUAT . Hsiao CC. Epstein-Barr virus associated with immune thrombocytopenic purpura in childhood: a retrospective study. J Paediatr Child Health. 2000;36(5):445–8. 10.1046/j.1440-1754.2000.00539.x . Zhang C, Kelly AM. Severe thrombocytopenia in a case of Epstein-Barr virus-induced infectious mononucleosis. Cureus. 2021;13(9):e17706. 10.7759/cureus.17706 . Published 2021 Sep 4. Yan M, Zhang Y, Yang F, Ji L, Wang M, Wang W. Comparative study between chronic immune thrombocytopenia patients and healthy population on Epstein-Barr virus infection status by polymerase chain reaction. Expert Rev Hematol. 2020;13(7):781–6. 10.1080/17474086.2020.1772746 . Li Y, Wang K. Clinical analysis of 163 pediatric patients with infectious mononucleosis: a single-center retrospective analysis. Immun Inflamm Dis. 2024;12(9):e70020. 10.1002/iid3.70020 . Han X, Xu P, Duan X, Liu Y, Zhang J, Xu H. High mean platelet volume-to-platelet count ratio as a diagnostic maker for increased risk of liver function damage in pediatric patients with infectious mononucleosis in China. Exp Ther Med. 2019;18(6):4523–7. 10.3892/etm.2019.8104 . Lin J, Chen X, Wu H, Chen X, Hu X, Xu J. Peripheral blood lymphocyte counts in patients with infectious mononucleosis or chronic active Epstein-Barr virus infection and prognostic risk factors of chronic active Epstein-Barr virus infection. Am J Transl Res. 2021;13(11):12797–806. Published 2021 Nov 15. Prtorić L, Šokota A, Karabatić Knezović S, Tešović G, Zidovec-Lepej S. Clinical features and laboratory findings of hospitalized children with infectious mononucleosis caused by Epstein-Barr virus from Croatia. Pathogens. 2025;14(4):374. Published 2025 Apr 10. 10.3390/pathogens14040374 Cabitza F, Rasoini R, Gensini GF. Unintended consequences of machine learning in medicine. JAMA. 2017;318(6):517–8. 10.1001/jama.2017.7797 . Lundberg SM, Lee SI. A unified approach to interpreting model predictions. In: NIPS'17: Proceedings of the 31st International Conference on Neural Information Processing Systems. 2017 Presented at: NIPS'17; Dec 4–9; Long Beach, CA pp. 4768–4777. Lundberg SM, Nair B, Vavilala MS, et al. Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. Nat Biomed Eng. 2018;2(10):749–60. 10.1038/s41551-018-0304-0 . Greener JG, Kandathil SM, Moffat L, Jones DT. A guide to machine learning for biologists. Nat Rev Mol Cell Biol. 2022;23(1):40–55. 10.1038/s41580-021-00407-0 . Wu L, Cao X, Wang J et al. Etiological stratification and prognostic assessment of haemophagocytic lymphohistiocytosis by machine learning on onco-mNGS data and clinical data. Front Immunol. 2024;15:1390298. Published 2024 Sep 9. 10.3389/fimmu.2024.1390298 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 15 Feb, 2026 Reviewers agreed at journal 05 Feb, 2026 Reviewers invited by journal 05 Feb, 2026 Editor assigned by journal 03 Feb, 2026 Editor invited by journal 12 Jan, 2026 Submission checks completed at journal 08 Jan, 2026 First submitted to journal 08 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8437947","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":588354448,"identity":"1aa0647b-34cc-4ac9-8112-8587721b79ae","order_by":0,"name":"Ruibing Zhao","email":"","orcid":"","institution":"Xi'an Children's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Ruibing","middleName":"","lastName":"Zhao","suffix":""},{"id":588354452,"identity":"e496b33e-244f-45db-83a1-0c4cef41e514","order_by":1,"name":"Ce Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1klEQVRIiWNgGAWjYPACCR559uYDBz5UEK/FRs6w51jiwRlniNeSZsxww8f4MG8LEWoNjp89/OJHxeHExhk8Hw7wNjDI84sdIKDlTF6aZc+Zw4nt0r0bDkjuYDCcOTsBvxazAzlmxoxtQFvmnN1wwPAMQ4LBbUJazr8Bavl3OLHhRs6DA4ltxGi5kWP8mLEB5P0chgMHidFif+ONGWPPMXAgGxxsOCNB2C+S/TnGH37UgKPy8ec/FTby/NIEtAABmwQSRwKnMmTA/IEoZaNgFIyCUTByAQB9gE++5cCBFQAAAABJRU5ErkJggg==","orcid":"","institution":"Xi'an Children's Hospital","correspondingAuthor":true,"prefix":"","firstName":"Ce","middleName":"","lastName":"Wang","suffix":""},{"id":588354453,"identity":"f6fb2012-13d7-47e2-a6a1-e4d672e58347","order_by":2,"name":"Qian Tian","email":"","orcid":"","institution":"Xi'an Children's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Qian","middleName":"","lastName":"Tian","suffix":""},{"id":588354454,"identity":"b529d6cc-111b-4747-8f1b-915256eb2066","order_by":3,"name":"Nan Wang","email":"","orcid":"","institution":"Xi'an Children's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Nan","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-12-24 02:08:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8437947/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8437947/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102376303,"identity":"9aaf6bcd-220b-4464-8426-bc8ae56c137c","added_by":"auto","created_at":"2026-02-11 05:27:00","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":245632,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLASSO regression analysis obtained 12 characteristic variables\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8437947/v1/074255cf723fd015a705cb50.png"},{"id":102398268,"identity":"53e339fa-1be7-4696-b841-f75d8762537f","added_by":"auto","created_at":"2026-02-11 10:21:59","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":167668,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe best clinical prediction model and AUC curve of PEBV-IM\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8437947/v1/b97dea12b81c9cc3ae602ba6.png"},{"id":102376304,"identity":"2928cf29-c2a1-472d-a034-25253e0b53ae","added_by":"auto","created_at":"2026-02-11 05:27:00","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":35175,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eInfluence of GGT and platelet count on model prediction.\u003c/strong\u003e A: Platelet count and GGT versus clinical predictive model | SHAP| values;B: Effect of platelet count and GGT on predictive model output; C: Correlation between platelet count and GGT.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8437947/v1/b5bf86549fe1d57e14ae648e.png"},{"id":102399123,"identity":"076aaacf-da34-42f9-b9dc-f5da798eedd3","added_by":"auto","created_at":"2026-02-11 10:33:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2023202,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8437947/v1/7cb0c284-721c-4bc4-810f-ccbf1804665b.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Clinical prediction models for pediatric Epstein-Barr virus infectious mononucleosis: from 6 machine learning algorithms Running Title: Machine learning Models for Pediatric EBV Infectious Mononucleosis","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eEpstein-Barr virus (EBV), also known as human herpesvirus 4(HHV-4), is a member of the gamma herpesvirus family that infects over 95% of the world's population\u003csup\u003e1\u003c/sup\u003e. EBV was first identified in Burkitt lymphoma (BL) in 1964 and was later found to be associated with other types of lymphoma, including Hodgkin lymphoma (HL), non-HL, T cell lymphoma, and natural killer (NK)/T cell lymphoma in posttransplant patients and HIV infected individuals. EBV is also linked to epithelial cancers, such as nasopharyngeal carcinoma (NPC) and gastric cancers.\u003c/p\u003e\n\u003cp\u003eAdditionally, EBV is associated with nonmalignant diseases, including infectious mononucleosis (IM), oral hairy leukoplakia, systemic lupus erythematosus (SLE), and multiple sclerosis (MS)\u003csup\u003e2\u003c/sup\u003e. Although several preventive or therapeutic vaccine strategies for EBV have been evaluated in clinical and preclinical trials, no licensed vaccines or therapeutic interventions for EBV-associated diseases\u003csup\u003e3\u0026ndash;5\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003ePrimary EBV infection is often asymptomatic, but when the immune response is strong, the resulting disease state is called IM. Patients with symptomatic IM typically experience the classic triad of fever, lymphadenopathy, and pharyngitis\u003csup\u003e6,7\u003c/sup\u003e, along with other symptoms such as transient hepatitis, splenomegaly, malaise, nausea, and palatal petechiae, depending on the host's response to the invading virus\u003csup\u003e8\u003c/sup\u003e. Painless bilateral swelling of the upper eyelids\u003csup\u003e9,10\u003c/sup\u003e and thrombocytopenia\u003csup\u003e11\u0026ndash;13\u003c/sup\u003e may also occur. Although IM is a self-limiting disease and symptoms usually resolve within weeks, fatigue may persist for months. Early, accurate laboratory results are essential for the correct diagnosis of IM, allowing for timely intervention and avoiding unnecessary treatments, such as antibiotics for pharyngitis symptoms or expensive exploratory tests for cases with splenomegaly or suspicious hematological disease\u003csup\u003e14\u003c/sup\u003e. Therefore, reliable diagnostic metrics and clinical predictive models are crucial for effective decision-making in PEBV-IM.\u003c/p\u003e\n\u003cp\u003eMachine Learning (ML) enables computer systems to learn from data, improve performance, and make predictions or decisions by identifying patterns and building mathematical models. Common methods used in clinical prediction models include linear regression, logistic regression, support vector machines, decision trees, random forests, neural networks. These methods can process large-scale and complex data, improve prediction accuracy, personalize treatment, and provide real-time decision-making tools for PEBV-IM.\u003c/p\u003e\n\u003cp\u003eBased on clinical and pathological data collected from PEBV-IM patients, this study constructs a clinical predictive model for PEBV-IM using bioinformatics analysis and multiple ML algorithms, identifying key variables related to PEBV-IM. This model provides a valuable decision-making tool for PEBV-IM, enabling early intervention and personalized treatment.\u003c/p\u003e"},{"header":"2. Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Data collection and processing\u003c/h2\u003e \u003cp\u003eData on clinical characteristics of pediatric Epstein-Barr virus (EBV) infectious mononucleosis (PEBV-IM) patients were collected from Xi'an Children's Hospital, and informed consent was obtained from the patient\u0026rsquo;s guardian for the used of the data. Firstly, the PEBV-IM dataset contained information on totally 99 cases with the disease, subsequently, variables with null data (days to EBV diagnosis, cluster of differentiation 25 (CD25), natural killer (NK) cell activity), variables with only one status (lymph node enlargement, history of immunosuppression or not), and variables with too large a disparity in sample size (sore throat / pharyngeal isthmus (yes: 97, no: 2), presence of haemophilia on bone puncture (not investigated: 95, yes: 2, no : 2)) were excluded. Finally, the 39 samples in which EBV load were not checked were excluded, a sum of 60 samples and 41 variables included in the analysis (\u003cb\u003eSupplementary Table\u0026nbsp;1\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Baseline characteristics analysis\u003c/h2\u003e \u003cp\u003eThe PEBV-IM dataset was first split into training and validation sets in a 7: 3 ratio by random splitting. Subsequently, so as to assess whether there were marked differences in the clinical feature variables between the training and validation sets. Subsequently, the quartile method\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e was adopted to split continuous variables into categorical variables (Q1-Q4), and the Mann-Whitney U test was applied to compare differences in clinical characteristics. Then, the chi-square test was applied to compare the differences in clinical characteristics of categorical variables, excluding variables with marked in the both dataset (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Construction of clinical prediction models\u003c/h2\u003e \u003cp\u003eTo identify feature variables for predicting PEBV-IM, founded on training set, the aspartate aminotransferase (AST) / alanine aminotransferase (ALT) was designated as the dependent variable (to be 0 for less than the median and 1 for greater than the median). Then, the glmnet package (v 4.1-8)\u003csup\u003e16\u003c/sup\u003e was adopted to proceed 10-fold cross-validation least absolute shrinkage and selection operator (LASSO) regression analysis. The variables with the smallest model error and the regression coefficient not penalized to 0 were selected as the feature variables. Moreover, founded on the feature variables, the NaiveBayes, RF, SVM, Ridge, plsRglm and LDA algorithms from e1071 package (v 1.7\u0026ndash;14),\u003csup\u003e17\u003c/sup\u003e randomForestSRC package (v 3.3.0)\u003csup\u003e18\u003c/sup\u003e, e1071 package (v 1.7\u0026ndash;14), glmnet package (v 4.1-8), plsRglm package (v 1.5.1)\u003csup\u003e19\u003c/sup\u003e, and MASS package (v 7.3\u0026ndash;60.0.1 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://cran.r-project.org/web/packages/MASS/index.html\u003c/span\u003e\u003cspan address=\"https://cran.r-project.org/web/packages/MASS/index.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) were applied to construct the clinical predictive models, respectively. Subsequently, in the training and validation sets, the pROC package (v 1.18.0)\u003csup\u003e20\u003c/sup\u003e was applied to evaluate the accuracy of predictive models by plotting receiver operating characteristic (ROC) curves, and the optimal clinical predictive model of PEBV-IM with the highest the area under curve (AUC) values in both datasets was selected. Further, founded on the optimal clinical prediction model, the key feature variables were acquired.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 SHapley Additive exPlanations (SHAP) model interpretation\u003c/h2\u003e \u003cp\u003eSo as to better understand the relationship between the key feature variables and the clinical prediction model, the shapviz package (v 0.9.5)\u003csup\u003e21\u003c/sup\u003e was applied to compute the mean |SHAP| values of the key feature variables and to acquire distribution of feature importance. Subsequently, the key feature variables were ranked in descending order of importance founded on the mean |SHAP| values of the key feature variables, the ggplot2 package (v 3.4.2)\u003csup\u003e22\u003c/sup\u003e was applied to visualize the results. Moreover, the impact of single key feature variables on the optimal clinical prediction model results was observed.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003e2.5 Statistical analysis\u003c/b\u003e\u003c/h2\u003e \u003cp\u003eAll statistical analyses were conducted using R software (v 4.3.1). The Mann-Whitney U test and chi-square test were utilized to evaluate paired sample comparisons, with statistical significance defined as p\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1 The 39 variables were applied in subsequent analyses\u003c/h2\u003e\n \u003cp\u003eA sum of 60 PEBV-IM samples and 41 variables were included in the study, with the training set and validation set containing 47 and 13 PEBV-IM samples, respectively. Then, the results of the baseline characteristics analysis displayed that the T cells and Ts cells count were markedly different in the training and validation sets (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), while the rest of the 39 variables were not markedly different (p\u0026thinsp;\u0026gt;\u0026thinsp;0.05) and could be applied for subsequent analyses (Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2 The optimal clinical predictive models and key characteristic variables were obtained\u003c/h2\u003e\n \u003cp\u003eIn the training set, founded on the 39 variables, 12 feature variables were acquired by LASSO regression analysis (log(lambda.min) = -2.9235 (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA, B), namely age, gender, degree of splenomegaly, percentage of lymphocytes, red blood cells count, hemoglobin, platelet count, GGT, helper T (Th) cells count, B cells count, interleukin (IL)-1 beta (IL-1\u0026beta;) and IL-10 (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). Subsequently, 6 clinical prediction models were acquired founded on the construction of 12 feature variables (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e), and it could be observed that the clinical prediction model constructed by plsRglm had the highest AUC values in the training (AUC\u0026thinsp;=\u0026thinsp;0.939) and validation sets (AUC\u0026thinsp;=\u0026thinsp;0.850) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eA-C), which was the optimal clinical predictive model for PEBV-IM. Furthermore, platelet count and GGT were identified as key feature variables.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eBaseline characterisation\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eLevel\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eValidation set\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eTraining set\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ep\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003en\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eAge (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.449\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" align=\"left\"\u003e\n \u003cp\u003eGender (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eman\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7 (53.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e25 (53.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" align=\"char\"\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ewoman\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6 (46.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22 (46.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eHighest body temperature (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.717\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6 (12.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6 (46.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16 (34.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eDays with body temperature\u0026thinsp;\u0026gt;\u0026thinsp;38℃ (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0 ( 0.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8 (17.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.318\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6 (12.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7 (53.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19 (40.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" align=\"left\"\u003e\n \u003cp\u003eDegree of liver enlargement (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6 (46.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20 (42.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\" align=\"char\"\u003e\n \u003cp\u003e0.964\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eModerate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7 (14.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNormal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20 (42.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" align=\"left\"\u003e\n \u003cp\u003eDegree of splenomegaly (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6 (12.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\" align=\"char\"\u003e\n \u003cp\u003e0.666\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eModerate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7 (53.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20 (42.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNormal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e21 (44.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eEpstein-Barr virus (EBV) load (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;5\u0026times;10\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6 (46.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e21 (44.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.513\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5\u0026times;10\u003csup\u003e3\u003c/sup\u003e~5\u0026times;10\u003csup\u003e4\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16 (34.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5\u0026times;10\u003csup\u003e4\u003c/sup\u003e~5\u0026times;10\u003csup\u003e5\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 ( 4.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5\u0026times;10\u003csup\u003e2\u003c/sup\u003e~5\u0026times;10\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8 (17.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eWhite blood cells count (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.601\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eNeutrophil count (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.782\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eLymphocyte count (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.300\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003ePercentage of lymphocytes (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.961\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cstrong\u003eContinued\u003c/strong\u003e Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e \u003cstrong\u003eBaseline characterisation\u003c/strong\u003e\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003cdiv class=\"colspec\" align=\"char\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Taba\" border=\"1\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eLevel\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eValidation set\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eTraining set\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ep\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eRed blood cells count (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.961\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eHemoglobin (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.637\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003ePlatelet count (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.601\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eTriglycerides (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.348\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eFibrinogen (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e9 (19.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.839\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eTotal bilirubin (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.449\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eDirect bilirubin (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.122\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6 (46.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e9 (19.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eAlanine aminotransferase (ALT) (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.782\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eAspartate aminotransferase (AST) (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.747\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eLactate dehydrogenase (LDH) (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.329\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cstrong\u003eContinued\u003c/strong\u003e Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e \u003cstrong\u003eBaseline characterisation\u003c/strong\u003e\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003cdiv class=\"colspec\" align=\"char\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Tabb\" border=\"1\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eLevel\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eValidation set\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eTraining set\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ep\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eGamma-glutamyl transferase (GGT) (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.446\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eAlbumin (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.601\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eFerritin (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.681\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eT cells count (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.021\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e7 (53.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e8 (17.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eSuppressor T (Ts) cells count (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.043\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6 (46.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e9 (19.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eHelper T (Th) cells count (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.601\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eNatural killer (NK) cells count (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.782\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eB cells count (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.329\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eInterleukin (IL) -1 beta (IL-1\u0026beta;) (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.259\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e9 (19.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e15 (31.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eIL-2 (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0 ( 0.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e8 (17.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.327\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e17 (36.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cstrong\u003eContinued\u003c/strong\u003e Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e \u003cstrong\u003eBaseline characterisation\u003c/strong\u003e\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption\u003e\u0026nbsp;\u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eVariables\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eLevel\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eValidation set\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eTraining set\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ep\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" align=\"left\"\u003e\n \u003cp\u003eIL-4 (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\" align=\"char\"\u003e\n \u003cp\u003e0.616\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 ( 4.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (76.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e32 (68.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" align=\"left\"\u003e\n \u003cp\u003eIL-5 (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (10.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\" align=\"char\"\u003e\n \u003cp\u003e0.637\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e9 (69.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e29 (61.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eIL-6 (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.727\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eIL-8 (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 ( 8.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.655\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e7 (53.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e17 (36.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eIL-10 (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.816\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" align=\"left\"\u003e\n \u003cp\u003eInterleukin-12 p70 subunit (IL-12p70) (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (10.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\" align=\"char\"\u003e\n \u003cp\u003e0.263\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (76.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e28 (59.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eIL-17 (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.304\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (10.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e8 (61.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e18 (38.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1 ( 7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" align=\"left\"\u003e\n \u003cp\u003eTumor necrosis factor-alpha (TNF-\u0026alpha;) (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e8 (17.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\" align=\"char\"\u003e\n \u003cp\u003e0.454\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e7 (53.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e26 (55.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eInterferon-gamma (IFN-\u0026gamma;) (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.782\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3 (23.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"4\" align=\"left\"\u003e\n \u003cp\u003eInterferon-alpha (IFN-\u0026alpha; )(%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"4\" align=\"char\"\u003e\n \u003cp\u003e0.145\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0 ( 0.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e14 (29.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5 (38.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10 (21.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eQ4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12 (25.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003cdiv class=\"colspec\" align=\"char\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 2 Characteristic Variable Analysis of PEBV-IM\u003c/strong\u003e\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv class=\"colspec\" align=\"char\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Tabc\" border=\"1\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eFeature variable\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCoefficient\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.02434\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.38919\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDegree of splenomegaly\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.14292\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePercentage of lymphocytes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.01561\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRed blood cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.50542\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ehemoglobin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.00313\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePlatelet count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.00491\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGGT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.01484\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTh cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.00005\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eB cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.00005\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIL-1\u0026beta;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.00874\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIL-10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.02497\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Tab3\" border=\"1\"\u003e\n \u003ccaption\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eMachine learning to acquire key feature variables\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eGGT\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eRF\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePlatelet count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ehemoglobin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRed blood cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eB cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRF\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDegree of splenomegaly\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePercentage of lymphocytes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRed blood cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ehemoglobin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePlatelet count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGGT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTh cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eB cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIL-1\u0026beta;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIL-10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNaiveBayes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDegree of splenomegaly\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePercentage of lymphocytes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRed blood cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ehemoglobin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePlatelet count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGGT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTh cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eB cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIL-1\u0026beta;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIL-10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDegree of splenomegaly\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePercentage of lymphocytes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cstrong\u003eContinued\u003c/strong\u003e Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e \u003cstrong\u003eMachine learning to acquire key feature variables\u003c/strong\u003e\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Tabd\" border=\"1\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eRed blood cells count\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ehemoglobin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePlatelet count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGGT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTh cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eB cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIL-1\u0026beta;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIL-10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRidge\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePlatelet count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eplsRglm\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGGT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eplsRglm\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDegree of splenomegaly\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePercentage of lymphocytes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRed blood cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ehemoglobin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePlatelet count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGGT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTh cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eB cells count\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIL-1\u0026beta;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIL-10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDA\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003e3.3 GGT and platelet count affected model output\u003c/h2\u003e\n \u003cp\u003eThe SHAP values could determine the difference between the predicted values for all combinations, by calculating the mean |SHAP| values, it could be observed that the GGT had the greatest effect on the clinical prediction model output (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eA). In addition, the effect of features on the model output was explored, the results displayed that samples with high GGT feature values had a negative effect on the prediction, and samples with low GGT feature values had a positive effect on the prediction. On the feature values of platelet count, a low feature values had a negative effect on the model output and conversely a positive effect (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eB). Finally, by focusing on the dependence between platelet count and GGT, it was found that both had a linear relationship with the model\u0026rsquo;s predictions. This indicated that the predictions were not only affected by the level of each key feature variable but also by the individual differences between them (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eC). Overall, the interpretability and transparency of the clinical prediction model was improved by SHAP interpretation, which enabled used to better understand the predictive results of the model for PEBV-IM risk prediction.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eEBV infection is a widespread condition that affects nearly all tissues and organs in the body\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, with systemic effects on the liver, blood and immune system in EBV-associated diseases\u003csup\u003e\u003cspan additionalcitationids=\"CR26\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Despite its prevalence, there is a critical lack of reliable predictive models for early identification and intervention in IM, particularly in pediatric populations. The present study addresses this gap by constructing a clinical predictive model using 6 ML algorithms, providing a powerful tool for early diagnosis and intervention in PEBV-IM.\u003c/p\u003e \u003cp\u003eMost primary EBV infections manifest as IM in the pediatric population, and in a few cases, they progress to more severe conditions like EBV-associated hemophagocytic lymphohistiocytosis (HLH)\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. Huang R et al.\u003csup\u003e29\u003c/sup\u003e used LASSO regression and logistic regression analysis to identify risk factors and independent predictors for EBV, while also constructing nomograms to assist in early detection. In addition, a small minority of individuals may develop chronically active Epstein-Barr virus (CAEBV) infection, after EBV infection, leading to a disease prone to relapse and associated with severe and potentially fatal comorbidities such as malignant lymphoma, HLH, multiple organ failure, and disseminated intravascular coagulation (DIC)\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eYuan JH et al.\u003csup\u003e32\u003c/sup\u003e enhanced their results by applying LASSO Logistic regression to identify key peripheral blood biomarkers that distinguish IM from CAEBV, thus offering valuable guidance for clinical work. Both studies highlight the reliability of LASSO as a high-dimensional predictor regression method\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. Similarly, in this study, 39 clinical variables were initially analyzed, with 12 characteristic variables identified through LASSO regression analysis. This led to the construction of 6 clinical predictive models, ultimately identifying GGT and platelet count as key characteristic variables for predicting PEBV-IM.\u003c/p\u003e \u003cp\u003eGGT is a glycosylated protein located on the outer plasma membrane, a microsomal enzyme involved in the homeostasis of glutathione and cysteine\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. GGT plays a crucial role in transferring glutamyl from gamma-glutamate to other peptides, and although its activity in the kidney is ten times higher than in the liver, serum GGT is typically used as a liver test\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. Studies have shown that about 85% of EBV-infected patients experience varying degrees of liver impairment, with some cases leading to hepatitis, which increases transaminase levels and impairs liver function\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. HeHC et al.\u003csup\u003e37\u003c/sup\u003e observed consistently elevated GGT levels in EBV-infected gastric cancer patients, while ZhangL\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e found GGT to be significantly higher in IM patients than in healthy controls, demonstrating its diagnostic and predictive value. These findings are consistent with our results, where high GGT characteristics were negatively associated with the prediction of IM, indicating that patients with higher GGT levels are more likely to be diagnosed with IM and experience longer hospital stays. Monitoring GGT can thus help help identify patients at greater risk and allow for early intervention.\u003c/p\u003e \u003cp\u003ePlatelet are essential for coagulation and the repair of vascular endothelial cells, with significant decreases in platelet count leading to spontaneous bleeding, which can be life-threatening in some severe cases\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. Previous studies have suggested that IM can cause thrombocytopenia in children\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. A recent study reported by Zhang et al described a 14-year-old girl who developed severe thrombocytopenia (with a platelet count down to 5\u0026times;10\u003csup\u003e9\u003c/sup\u003e/L), leading to spontaneous bleeding and periorbital edema, a rare symptom of IM\u003csup\u003e41\u003c/sup\u003e. Yan et al. reported EBV-DNA positivity in blood lymphocytes of patients with chronic immune thrombocytopenia (detected by quantitative reverse transcription PCR) higher than controls\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. LiY et al.\u003csup\u003e43\u003c/sup\u003e found in a study that EBV-DNA copy number in plasma can be used as an important indicator to assess the duration of fever and clinical outcomes in children. These researches showed that platelet counts correlate with EBV-DNA copy number in plasma, suggesting that this indicator can be used to assess the duration of fever and clinical outcomes in children. HanX et al.\u003csup\u003e44\u003c/sup\u003e used regression analysis to analyze the predictive value of platelet four indicators in IM patients, and established three regression models for regression analysis, showing that PLT is a significant predictor of IM. LinJ et al.\u003csup\u003e45\u003c/sup\u003e proposed in their study that L and platelet count\u0026thinsp;\u0026lt;\u0026thinsp;50 \u0026times; 10\u003csup\u003e12\u003c/sup\u003e/L are independent risk factors for mortality in CAEBV patients. PrtorićL et al.\u003csup\u003e46\u003c/sup\u003e reported in a recent article that each unit decrease in platelet count increases hospital stay by 0.12 days, so platelet count is a predictor of hospital stay. Our study supports these findings, with low platelet eigenvalues negatively impacting the model output, and vice versa. That is, the lower the platelet number, the more severe the child's condition and the longer the hospital stay. Therefore, platelet count can be used to clinically assess the severity of IM in children, helping clinicians identify those at risk and intervene early.\u003c/p\u003e \u003cp\u003eIn this study, we developed the best clinical predictive model for PEBV-IM through 6 ML methods, particularly the plsRglm model. However, a major challenge in the medical application of ML models is the lack of interpretability\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. To address this, we introduced SHAP values, which were proposed by Lundberg et al.\u003csup\u003e48\u003c/sup\u003e for accurately calculating the contribution and impact (positive or negative) of each feature on the final prediction. SHAP can perform both local and global interpretability and has a solid theoretical foundation compared to other methods\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. In this study, SHAP values allowed us to better understand the impact of key characteristic variables (GGT and platelet count) on the model output. Interestingly, the relationship between GGT and platelet count was linear, with higher GGT characteristic value and lower platelet characteristic value being associated with more severe IM. This further emphasizes that predictions were influenced not only by the level of each key characteristic variable but also by their individual differences between them.\u003c/p\u003e \u003cp\u003eML, especially in healthcare, holds great promise in handling large, complex datasets to improve predictions. Greener JG\u003csup\u003e50\u003c/sup\u003e highlighted the importance of matching predictive models to data and automating analysis to build repeatable and time-saving pipelines. ML algorithms are particularly useful when dealing with vast amounts of clinical data, as demonstrated by their widespread application in discovering diagnostic markers and predicting disease outcomes\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e. Our study highlights the potential of ML in clinical settings, particularly for identifying key biomarkers and predicting disease progression in IM patients. This could ultimately lead to more personalized and effective treatments.\u003c/p\u003e \u003cp\u003eDespite the promising results, there are several limitations in this study that warrant consideration. First, the sample size is relatively small, which may limit the generalizability and robustness of the model. Larger and multi-center studies are needed to validate the findings and enhance the model's applicability across diverse clinical settings. Second, some variables were missing in the dataset, which could introduce bias into the analysis. Lastly, while ML models are powerful tools, their complexity and the difficulty in interpreting results may hinder their repeatability and widespread use in clinical practice. Future research should focus on improving the model\u0026rsquo;s interpretability, expanding the clinical data set, and conducting clinical trials to verify its real-world applicability.\u003c/p\u003e \u003cp\u003eIn conclusion, our study demonstrates the potential of ML in creating an effective predictive model for PEBV-IM. The identification of key characteristic variables (GGT and platelet count) provides valuable insights for early diagnosis and intervention. By integrating these predictive tools into clinical practice, we can improve patient outcomes through more personalized and timely interventions. Future studies should aim to refine and validate these models for broader clinical use.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eabbreviation\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003efull name\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEBV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eEpstein-Barr virus\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePEBV-IM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePediatric Epstein-Barr virus Infectious mononucleosis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLASSO\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eleast absolute shrinkage and selection operator\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eML\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003emachine learning\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSHAP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSHapley Additive exPlanations\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAUC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003earea under the curve\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eGGT\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003egamma-glutamyl transferase\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eHHV-4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ehuman herpes virus 4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBurkitt lymphoma\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eHL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eHodgkin lymphoma\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNK\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003enatural killer\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eNPC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003enasopharyngeal carcinoma\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eIM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003einfectious mononucleosis\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSLE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003esystemic lupus erythematosus\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003emultiple sclerosis\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eML\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMachine Learning\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eCD25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ecluster of differentiation 25\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003easpartate aminotransferase\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eALT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ealanine aminotransferase\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eROC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ereceiver operating characteristic\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eTh\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ehelper T\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLDH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLactate dehydrogenase\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eTs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSuppressor T\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eIL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eInterleukin\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eIL-1β\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eInterleukin -1 beta\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eIL-12p70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eInterleukin-12 p70 subunit\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eTNF-α\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eTumor necrosis factor-alpha\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eIFN-γ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eInterferon-gamma\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eIFN-α\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eInterferon-alpha\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eHLH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ehemophagocytic lymphohistiocytosis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eCAEBV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003echronically active Epstein-Barr virus\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eDIC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003edisseminated intravascular coagulation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eEthics approval and consent to participate\u003c/h2\u003e \u003cp\u003eThe study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethical Committee for Drug Clinical Trials at Xi 'an Children's Hospital (Ethics Approval No.: 20260036). Informed consent has been obtained from the patient.For the minors under 16 years old in this study, written informed consent has been obtained from their parents (or legal guardians).\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent for publication\u003c/strong\u003e \u003cp\u003eNot applicable\u003c/p\u003e \u003c/p\u003e\u003cp\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors declare that they have no competing interests\u003c/p\u003e \u003c/p\u003e\u003cp\u003e \u003ch2\u003eClinical trial number\u003c/h2\u003e \u003cp\u003eNot applicable\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eNot applicable\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eCeWang:Writing-review\u0026amp;amp; editing, Conceptualization, Methodology, Project administration;RuibingZhao: Writing \u0026ndash; review \u0026amp;amp; editing, Conceptualization, Methodology; CeWang: Writing \u0026ndash; original draft, Conceptualization, Methodology; RuibingZhao: Writing \u0026ndash; review \u0026amp;amp; editing, Formal Analysis; QianTian: Writing \u0026ndash; review \u0026amp;amp; editing; Nan Wang: Writing \u0026ndash; review \u0026amp;amp; editing\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eWe would like to express our sincere gratitude to all individuals and organizations who supported and assisted us throughout this research. Special thanks to the following authors: In conclusion, we extend our thanks to everyone who has supported and assisted us along the way. Without your support, this research would not have been possible.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets generated and/or analyzed during the current study are not publicly available due to the protection of patients' privacy, but can be obtained from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eDamania B, Kenney SC, Raab-Traub N. Epstein-Barr virus: Biology and clinical disease. Cell. 2022;185(20):3652\u0026ndash;70. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2022.08.026\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2022.08.026\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGewurz BL, Longnecker RM, Cohen JI. (2021). Epstein-Barr virus(Chap. 11). Fields Virology, 324\u0026ndash;388.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmith C, Khanna R. The development of prophylactic and therapeutic EBV vaccines. Curr Top Microbiol Immunol. 2015;391:455\u0026ndash;73. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/978-3-319-22834-1_16\u003c/span\u003e\u003cspan address=\"10.1007/978-3-319-22834-1_16\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCohen JI. Vaccine development for Epstein-Barr virus. Adv Exp Med Biol. 2018;1045:477\u0026ndash;93. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/978-981-10-7230-7_22\u003c/span\u003e\u003cspan address=\"10.1007/978-981-10-7230-7_22\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDasari V, Bhatt KH, Smith C, Khanna R. Designing an effective vaccine to prevent Epstein-Barr virus-associated diseases: challenges and opportunities. Expert Rev Vaccines. 2017;16(4):377\u0026ndash;90. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/14760584.2017.1293529\u003c/span\u003e\u003cspan address=\"10.1080/14760584.2017.1293529\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHOAGLAND RJ. The clinical manifestations of infectious mononucleosis: a report of two hundred cases. Am J Med Sci. 1960;240:55\u0026ndash;63.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMing Y, Cheng S, Chen Z, et al. Infectious mononucleosis in children and differences in biomarker levels and other features between disease caused by Epstein-Barr virus and other pathogens: a single-center retrospective study in China. PeerJ. 2023;11:e15071. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7717/peerj.15071\u003c/span\u003e\u003cspan address=\"10.7717/peerj.15071\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Published 2023 Apr 6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao Y, Li L, Hu X, Zhang W, Li Y. Interleukin-35 has a protective role in infectious mononucleosis-induced liver inflammation probably by inhibiting CD8\u0026thinsp;+\u0026thinsp;T cell function. Arch Immunol Ther Exp (Warsz). 2022;70(1):25. Published 2022 Oct 11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00005-022-00663-8\u003c/span\u003e\u003cspan address=\"10.1007/s00005-022-00663-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRicardo D. A Protracted Course of Periorbital Oedema in Infectious Mononucleosis Caused by Epstein-Barr Virus. Infect Dis Rep. 2022;14(6):942\u0026ndash;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/idr14060092\u003c/span\u003e\u003cspan address=\"10.3390/idr14060092\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Published 2022 Nov 23.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBronz G, Zanetti BPESM, Bianchetti MG, et al. Bilateral upper eyelid swelling (Hoagland sign) in Epstein-Barr infectious mononucleosis: prospective experience. Infection. 2023;51(2):471\u0026ndash;4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s15010-022-01932-6\u003c/span\u003e\u003cspan address=\"10.1007/s15010-022-01932-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eP\u0026aacute;ez-Guill\u0026aacute;n EM, Campos-Franco J, Alende R, Gonzalez-Quintela A. Hematological abnormalities beyond lymphocytosis during infectious mononucleosis: Epstein-Barr virus-induced thrombocytopenia. Mediterr J Hematol Infect Dis. 2023;15(1):e2023023. Published 2023 Mar 1. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.4084/MJHID.2023.023\u003c/span\u003e\u003cspan address=\"10.4084/MJHID.2023.023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang C, Kelly AM. Severe Thrombocytopenia in a case of Epstein-Barr virus-induced infectious mononucleosis. Cureus. 2021;13(9):e17706. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7759/cureus.17706\u003c/span\u003e\u003cspan address=\"10.7759/cureus.17706\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Published 2021 Sep 4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYusuf H, Kou A, Zelinskas C, et al. Secondary immune thrombocytopenic purpura due to primary Epstein- Barr virus infection. Cureus. 2022;14(6):e26112. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7759/cureus.26112\u003c/span\u003e\u003cspan address=\"10.7759/cureus.26112\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Published 2022 Jun 20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaughton P, Healy M, Enright F, Lucey B. Infectious Mononucleosis: diagnosis and clinical interpretation. Br J Biomed Sci. 2021;78(3):107\u0026ndash;16. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/09674845.2021.1903683\u003c/span\u003e\u003cspan address=\"10.1080/09674845.2021.1903683\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRuilope LM, Agarwal R, Anker SD, et al. Blood Pressure and Cardiorenal Outcomes With Finerenone in Chronic Kidney Disease in Type 2 Diabetes. Hypertension. 2022;79(12):2685\u0026ndash;95. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1161/HYPERTENSIONAHA.122.19744\u003c/span\u003e\u003cspan address=\"10.1161/HYPERTENSIONAHA.122.19744\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFriedman J, Hastie T, Tibshirani R. Regularization Paths for Generalized Linear Models via Coordinate Descent. J Stat Softw. 2010;33(1):1\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCinelli M, Sun Y, Best K, et al. Feature selection using a one dimensional na\u0026iuml;ve Bayes' classifier increases the accuracy of support vector machine classification of CDR3 repertoires. Bioinformatics. 2017;33(7):951\u0026ndash;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/bioinformatics/btw771\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/btw771\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao P, Zhen H, Zhao H, Huang Y, Cao B. Identification of hub genes and potential molecular mechanisms related to radiotherapy sensitivity in rectal cancer based on multiple datasets. J Transl Med. 2023;21(1):176. Published 2023 Mar 6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12967-023-04029-2\u003c/span\u003e\u003cspan address=\"10.1186/s12967-023-04029-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYin Y, Chen C, Zhang D et al. Construction of predictive model of interstitial fibrosis and tubular atrophy after kidney transplantation with machine learning algorithms. Front Genet. 2023;14:1276963. Published 2023 Nov 1. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fgene.2023.1276963\u003c/span\u003e\u003cspan address=\"10.3389/fgene.2023.1276963\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobin X, Turck N, Hainard A, et al. pROC: an open-source package for R and S\u0026thinsp;+\u0026thinsp;to analyze and compare ROC curves. BMC Bioinformatics. 2011;12:77. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/1471-2105-12-77\u003c/span\u003e\u003cspan address=\"10.1186/1471-2105-12-77\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Published 2011 Mar 17.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang TF, Luo C, Guo LB, et al. Preoperative prediction of textbook outcome in intrahepatic cholangiocarcinoma by interpretable machine learning: A multicenter cohort study. World J Gastroenterol. 2025;31(11):100911. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3748/wjg.v31.i11.100911\u003c/span\u003e\u003cspan address=\"10.3748/wjg.v31.i11.100911\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGustavsson EK, Zhang D, Reynolds RH, Garcia-Ruiz S, Ryten M. ggtranscript: an R package for the visualization and interpretation of transcript isoforms using ggplot2. Bioinformatics. 2022;38(15):3844\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/bioinformatics/btac409\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/btac409\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHouen G, Trier NH. Epstein-Barr Virus and systemic autoimmune diseases. Front Immunol. 2021;11:587380. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fimmu.2020.587380\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2020.587380\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Published 2021 Jan 7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFugl A, Andersen CL. Epstein-Barr virus and its association with disease - a review of relevance to general practice. BMC Fam Pract. 2019;20(1):62. Published 2019 May 14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12875-019-0954-3\u003c/span\u003e\u003cspan address=\"10.1186/s12875-019-0954-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang C, Cui S, Mao G, Li G. Clinical characteristics and the risk factors of hepatic injury in 221 children with infectious mononucleosis. Front Pediatr. 2022;9:809005. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fped.2021.809005\u003c/span\u003e\u003cspan address=\"10.3389/fped.2021.809005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Published 2022 Jan 12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVine LJ, Shepherd K, Hunter JG, et al. Characteristics of Epstein-Barr virus hepatitis among patients with jaundice or acute hepatitis. Aliment Pharmacol Ther. 2012;36(1):16\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/j.1365-2036.2012.05122.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1365-2036.2012.05122.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang SI, Geong JH, Kim JY. Clinical characteristics of primary Epstein Barr virus hepatitis with elevation of alkaline phosphatase and γ-glutamyltransferase in children. Yonsei Med J. 2014;55(1):107\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3349/ymj.2014.55.1.107\u003c/span\u003e\u003cspan address=\"10.3349/ymj.2014.55.1.107\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaylor GS, Long HM, Brooks JM, Rickinson AB, Hislop AD. The immunology of Epstein-Barr virus-induced disease. Annu Rev Immunol. 2015;33:787\u0026ndash;821. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1146/annurev-immunol-032414-112326\u003c/span\u003e\u003cspan address=\"10.1146/annurev-immunol-032414-112326\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang R, Wu D, Wang L, et al. A predictive model for Epstein-Barr virus-associated hemophagocytic lymphohistiocytosis. Front Immunol. 2024;15:1503118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fimmu.2024.1503118\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2024.1503118\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Published 2024 Dec 5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaughton P, Healy M, Enright F, Lucey B. Infectious Mononucleosis: diagnosis and clinical interpretation. Br J Biomed Sci. 2021;78(3):107\u0026ndash;16. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/09674845.2021.1903683\u003c/span\u003e\u003cspan address=\"10.1080/09674845.2021.1903683\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim WY, Montes-Mojarro IA, Fend F, Quintanilla-Martinez L. Epstein-Barr virus-associated T and NK-cell lymphoproliferative diseases. Front Pediatr. 2019;7:71. Published 2019 Mar 15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fped.2019.00071\u003c/span\u003e\u003cspan address=\"10.3389/fped.2019.00071\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYuan JH, Pang CJ, Yuan SL. Constructing a predictive model based on peripheral blood signs to differentiate infectious mononucleosis from chronic active EBV infection. J Infect Dev Ctries. 2024;18(9):1429\u0026ndash;34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3855/jidc.19233\u003c/span\u003e\u003cspan address=\"10.3855/jidc.19233\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Published 2024 Sep 30.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang G, Qi L, Sun Z, et al. Evaluation and analysis of incidence and risk factors of lower extremity venous thrombosis after urologic surgeries: A prospective two-center cohort study using LASSO-logistic regression. Int J Surg. 2021;89:105948. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ijsu.2021.105948\u003c/span\u003e\u003cspan address=\"10.1016/j.ijsu.2021.105948\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang S, Jiang D, Tao Y. Role of gamma-glutamyltransferase in cardiovascular diseases. Exp Clin Cardiol. 2013;18(1):53\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSotil EU, Jensen DM. Serum enzymes associated with cholestasis. Clin Liver Dis. 2004;8(1):41\u0026ndash;vi. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/S1089-3261(03)00136-3\u003c/span\u003e\u003cspan address=\"10.1016/S1089-3261(03)00136-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMecadon K, Jandovitz N, Salerno D, Martinez M, Kato T, Sammons C. Treatment of Epstein\u0026ndash;Barr virus viremia in pediatric intestinal and liver transplant recipients. Transplantation. 2017;101:S135.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe HC, Han R, Xu BH et al. Circulating Epstein-Barr virus DNA associated with hepatic impairment and its diagnostic and prognostic role in patients with gastric cancer. Front Med (Lausanne). 2022;9:1028033. Published 2022 Oct 6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fmed.2022.1028033\u003c/span\u003e\u003cspan address=\"10.3389/fmed.2022.1028033\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang L, Zhou P, Meng Z, et al. Infectious mononucleosis and hepatic function. Exp Ther Med. 2018;15(3):2901\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3892/etm.2018.5736\u003c/span\u003e\u003cspan address=\"10.3892/etm.2018.5736\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRosado FG, Kim AS. Hemophagocytic lymphohistiocytosis: an update on diagnosis and pathogenesis. Am J Clin Pathol. 2013;139(6):713\u0026ndash;27. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1309/AJCP4ZDKJ4ICOUAT\u003c/span\u003e\u003cspan address=\"10.1309/AJCP4ZDKJ4ICOUAT\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHsiao CC. Epstein-Barr virus associated with immune thrombocytopenic purpura in childhood: a retrospective study. J Paediatr Child Health. 2000;36(5):445\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1046/j.1440-1754.2000.00539.x\u003c/span\u003e\u003cspan address=\"10.1046/j.1440-1754.2000.00539.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang C, Kelly AM. Severe thrombocytopenia in a case of Epstein-Barr virus-induced infectious mononucleosis. Cureus. 2021;13(9):e17706. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7759/cureus.17706\u003c/span\u003e\u003cspan address=\"10.7759/cureus.17706\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Published 2021 Sep 4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan M, Zhang Y, Yang F, Ji L, Wang M, Wang W. Comparative study between chronic immune thrombocytopenia patients and healthy population on Epstein-Barr virus infection status by polymerase chain reaction. Expert Rev Hematol. 2020;13(7):781\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/17474086.2020.1772746\u003c/span\u003e\u003cspan address=\"10.1080/17474086.2020.1772746\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Wang K. Clinical analysis of 163 pediatric patients with infectious mononucleosis: a single-center retrospective analysis. Immun Inflamm Dis. 2024;12(9):e70020. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/iid3.70020\u003c/span\u003e\u003cspan address=\"10.1002/iid3.70020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan X, Xu P, Duan X, Liu Y, Zhang J, Xu H. High mean platelet volume-to-platelet count ratio as a diagnostic maker for increased risk of liver function damage in pediatric patients with infectious mononucleosis in China. Exp Ther Med. 2019;18(6):4523\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3892/etm.2019.8104\u003c/span\u003e\u003cspan address=\"10.3892/etm.2019.8104\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin J, Chen X, Wu H, Chen X, Hu X, Xu J. Peripheral blood lymphocyte counts in patients with infectious mononucleosis or chronic active Epstein-Barr virus infection and prognostic risk factors of chronic active Epstein-Barr virus infection. Am J Transl Res. 2021;13(11):12797\u0026ndash;806. Published 2021 Nov 15.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePrtorić L, Šokota A, Karabatić Knezović S, Tešović G, Zidovec-Lepej S. Clinical features and laboratory findings of hospitalized children with infectious mononucleosis caused by Epstein-Barr virus from Croatia. Pathogens. 2025;14(4):374. Published 2025 Apr 10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/pathogens14040374\u003c/span\u003e\u003cspan address=\"10.3390/pathogens14040374\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCabitza F, Rasoini R, Gensini GF. Unintended consequences of machine learning in medicine. JAMA. 2017;318(6):517\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jama.2017.7797\u003c/span\u003e\u003cspan address=\"10.1001/jama.2017.7797\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLundberg SM, Lee SI. A unified approach to interpreting model predictions. In: NIPS'17: Proceedings of the 31st International Conference on Neural Information Processing Systems. 2017 Presented at: NIPS'17; Dec 4\u0026ndash;9; Long Beach, CA pp. 4768\u0026ndash;4777.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLundberg SM, Nair B, Vavilala MS, et al. Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. Nat Biomed Eng. 2018;2(10):749\u0026ndash;60. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41551-018-0304-0\u003c/span\u003e\u003cspan address=\"10.1038/s41551-018-0304-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreener JG, Kandathil SM, Moffat L, Jones DT. A guide to machine learning for biologists. Nat Rev Mol Cell Biol. 2022;23(1):40\u0026ndash;55. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41580-021-00407-0\u003c/span\u003e\u003cspan address=\"10.1038/s41580-021-00407-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu L, Cao X, Wang J et al. Etiological stratification and prognostic assessment of haemophagocytic lymphohistiocytosis by machine learning on onco-mNGS data and clinical data. Front Immunol. 2024;15:1390298. Published 2024 Sep 9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fimmu.2024.1390298\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2024.1390298\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-informatics-and-decision-making","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"midm","sideBox":"Learn more about [BMC Medical Informatics and Decision Making](http://bmcmedinformdecismak.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/midm/default.aspx","title":"BMC Medical Informatics and Decision Making","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Epstein-Barr virus, Infectious mononucleosis, Pediatric Epstein-Barr virus infectious mononucleosis, Clinical prediction model, Machine learning, SHAP","lastPublishedDoi":"10.21203/rs.3.rs-8437947/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8437947/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003ePediatric Epstein-Barr virus (EBV) Infectious mononucleosis (PEBV-IM) is an acute infectious disease. However, there are no effective clinical diagnostic indicators for PEBV-IM. The study aimed to construct a clinical model for effective prediction of PEBV-IM and to identify relevant key feature variables, thereby providing a favorable clinical decision-making tool for PEBV-IM.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eData were obtained from the clinical diagnosis of PEBV-IM patients, and the feature variables were acquired by least absolute shrinkage and selection operator (LASSO) regression analysis. Subsequently, optimal clinical prediction models and key feature variables for PEBV-IM were acquired using 6 machine learning (ML) algorithms. Finally, to further investigate the relationship between optimal clinical prediction models and key feature variables, the SHapley Additive exPlanations (SHAP) model interpretation was proceeded.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eA total of 60 PEBV-IM samples and 41 variables were included in the analyses, and 12 characteristic feature variables were identified by LASSO. Subsequently, founded on the feature variables, the clinical prediction model was constructed using the plsRglm algorithm, which achieved the highest accuracy in both the training set (area under the curve (AUC)\u0026thinsp;=\u0026thinsp;0.939) and the validation set (AUC\u0026thinsp;=\u0026thinsp;0.850). Thus, this model was identified as the optimal clinical prediction model, while key feature variables platelet count and gamma-glutamyl transferase (GGT) were acquired. Notably, the GGT had the significant effect on the output of the clinical prediction model, with low GGT having a positive effect on the output, while low feature values of platelet count had a negative effect on the model output.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eObtaining a highly accurate clinical prediction model for PEBV-IM and 2 key feature variables (platelet count and GGT), which, in combination with SHAP model interpretation, provided a clear understanding and a novel tool for early diagnosis and clinical decision-making in PEBV-IM.\u003c/p\u003e","manuscriptTitle":"Clinical prediction models for pediatric Epstein-Barr virus infectious mononucleosis: from 6 machine learning algorithms Running Title: Machine learning Models for Pediatric EBV Infectious Mononucleosis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-11 05:25:36","doi":"10.21203/rs.3.rs-8437947/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-02-16T02:13:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"230630741464645983441209498889362369550","date":"2026-02-05T23:28:07+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-05T20:52:51+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-03T06:56:32+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-01-12T16:46:44+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-09T03:42:21+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Medical Informatics and Decision Making","date":"2026-01-09T03:35:52+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-informatics-and-decision-making","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"midm","sideBox":"Learn more about [BMC Medical Informatics and Decision Making](http://bmcmedinformdecismak.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/midm/default.aspx","title":"BMC Medical Informatics and Decision Making","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f632ae35-01ef-432f-b018-8a346b2c60fd","owner":[],"postedDate":"February 11th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-02-11T05:25:36+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-11 05:25:36","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8437947","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8437947","identity":"rs-8437947","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}