Development and Validation of a Screening Model for Early Diagnosis of Biliary Atresia in Neonates with Cholestasis

Development and Validation of a Screening Model for Early Diagnosis of Biliary Atresia in Neonates with Cholestasis | 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 Development and Validation of a Screening Model for Early Diagnosis of Biliary Atresia in Neonates with Cholestasis Zhaozhou Liu, Yuyan Jin, Yong Zhao, Yanan Zhang, Shuangshuang Li, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7460903/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 20 Dec, 2025 Read the published version in Pediatric Surgery International → Version 1 posted 7 You are reading this latest preprint version Abstract Background Biliary atresia (BA) is a progressive neonatal cholestatic liver disease that requires timely diagnosis and intervention. Differentiating BA from other causes of neonatal cholestasis remains a significant clinical challenge. Methods In this study, we retrospectively analyzed the clinical and biochemical data of 243 cholestatic neonates, comprising 61 with BA and 182 with non-BA. We utilized five supervised machine learning algorithms—logistic regression (LRM), decision tree (DET), multilayer perceptron (MLP), support vector machine (SVC), and random forest (RF)—to construct diagnostic models for BA. The performance of each model was evaluated based on its accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). We then developed an online diagnostic tool based on the best-performing model. Results The BA and non-BA groups showed significant differences across multiple biochemical markers. All five models demonstrated good diagnostic performance, with the random forest (RF) model achieving the best results (AUC = 0.93, sensitivity = 88.5%, specificity = 85.2%). The combination of multiple biochemical parameters substantially improved diagnostic accuracy compared to using single indicators. The web-based tool provides an intuitive and user-friendly interface to support early BA screening in clinical practice. Conclusion Machine learning-based models, particularly the RF model, show great potential for the early diagnosis of BA in cholestatic neonates. The implementation of a dedicated online platform may facilitate timely identification and assist clinicians in decision-making. Biliary atresia Neonate Diagnosis Machine learning Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Introduction Biliary atresia (BA) is a severe neonatal cholestatic liver disease that causes progressive fibrosis and obstruction of both intrahepatic and extrahepatic bile ducts. The disease affects approximately 1 in 5,000 to 20,000 live births, with the highest incidence reported in Asian populations 1 . Without treatment, BA rapidly progresses to cirrhosis and liver failure, and it represents one of the leading indications for liver transplantation in infants 2 . Kasai portoenterostomy is the standard surgical treatment for BA. Early diagnosis followed by timely Kasai surgery is essential for improving outcomes because it helps preserve native liver function and may delay or even prevent liver transplantation 3 . Our previous work demonstrated that infants who undergo Kasai surgery during the neonatal period achieve higher rates of postoperative jaundice clearance and native liver survival 4 . However, early diagnosis of BA remains challenging in clinical practice. Clinical manifestations such as persistent jaundice and acholic stools often appear early but are nonspecific and overlap with other causes of neonatal cholestasis. Diagnostic imaging techniques, including abdominal ultrasound, hepatobiliary scintigraphy, and magnetic resonance cholangiopancreatography, are widely used, yet their performance depends heavily on operator expertise and disease stage 5 – 7 . Liver biopsy may provide histological support but is invasive and may lack specificity in early-stage disease 8 . To improve early recognition, recent studies have evaluated serum biochemical markers. Conventional parameters such as gamma-glutamyl transferase (GGT), total bilirubin, direct bilirubin, and total bile acids are helpful in differentiating BA from non-BA cholestasis 9 – 11 . In addition, novel biomarkers such as matrix metalloproteinase-7 (MMP-7) and amyloid-beta show promising diagnostic value 12 – 14 . However, their widespread clinical use remains restricted because of high cost, technical complexity, and limited availability. By contrast, routine biochemical tests are inexpensive, universally available, and already integrated into neonatal care, making them an ideal basis for developing practical diagnostic screening models. Machine learning (ML) techniques have gained increasing importance in medical diagnostics. These methods can detect complex data patterns and identify nonlinear relationships that conventional approaches often miss. In the context of BA, ML-based models have integrated symptoms, laboratory data, and imaging findings to improve diagnostic accuracy 7 , 15 – 18 . However, most existing studies focus on mixed-age pediatric populations, and few specifically address ML-assisted early diagnosis of neonatal BA. Moreover, most models have not yet been adapted for practical use at the bedside. Therefore, the present study aimed to develop and validate a machine learning-based diagnostic screening model for neonatal BA using routine serum biochemical indicators and basic demographic data. Furthermore, we deployed the optimal model as a web-based interactive platform to enable rapid, bedside clinical application and to facilitate early identification of infants at risk for BA. 2. Materials and Methods 2.1 Patient Selection and Data Collection This retrospective cohort study included cholestatic neonates evaluated at Beijing Children’s Hospital, Capital Medical University, between January 2017 and December 2023. Cholestasis was defined according to established biochemical criteria: (1) direct/conjugated bilirubin > 17.1 µmol/L when total bilirubin (TBIL) ≤ 85.5 µmol/L, or (2) direct/conjugated bilirubin fraction > 20% of TBIL when TBIL > 85.5 µmol/L. The diagnosis of BA was confirmed by intraoperative cholangiography. Non-BA etiologies were determined using a combination of intraoperative cholangiography, liver histopathology, genetic testing, and clinical response to conservative treatment. Eligible participants were neonates aged ≤ 30 days who presented with biochemically confirmed cholestasis. Exclusion criteria included incomplete diagnostic records, indeterminate final diagnosis, or lack of follow-up at our institution. Ethical approval was obtained from the Institutional Review Board of Beijing Children’s Hospital (Approval No. 2019-k-386), with informed consent waived because of the retrospective and anonymized study design. 2.2 Data Collection and Biochemical Parameters Comprehensive demographic characteristics and baseline laboratory parameters were extracted from the electronic medical records during the first clinical encounter prior to therapeutic intervention. The complete biochemical panel comprised 37 parameters across three physiological domains: (1) Electrolytes and renal function markers including potassium (K), sodium (Na), chloride (Cl), carbon dioxide (CO₂), anion gap, osmolality, creatinine, urea, and urea-to-creatinine ratio; (2) Hepatic metabolic markers encompassing total protein, albumin, globulin, albumin-to-globulin ratio, total cholesterol, triglycerides, glucose, calcium, phosphorus, magnesium, uric acid, and lipid fractions (HDL-C, LDL-C, VLDL-C); and (3) Cholestasis-specific biomarkers such as total bilirubin, direct bilirubin, indirect bilirubin, total bile acids, alkaline phosphatase, γ-glutamyl transferase, aspartate aminotransferase, alanine aminotransferase, AST-to-ALT ratio, prealbumin, creatine kinase, and lactate dehydrogenase. 2.3 Development and Validation of Machine Learning Algorithms We applied five supervised machine learning algorithms to construct diagnostic screening models for neonatal BA. These algorithms included logistic regression model (LRM), decision tree (DET), multilayer perceptron (MLP), support vector machine classifier (SVC), and random forest (RF). All analyses were performed in Python 3.10 using the scikit-learn (v1.2.2) library. The cohort of 243 neonates with cholestasis (61 BA, 182 non-BA) was randomly partitioned into a training set (70%, n = 170) and a validation set (30%, n = 73) using stratified sampling to balance age and sex distributions. Feature engineering comprised multicollinearity reduction through exclusion of highly correlated variables (|Spearman's r| >0.90), followed by feature selection using L1-penalized LASSO regression. Each model underwent rigorous optimization through 10-fold cross-validation and hyperparameter tuning via Optuna framework. Performance was evaluated in validation cohorts using six core metrics: area under receiver operating characteristic curve (AUC-ROC), accuracy, precision, recall, F1-score, and calibration curve analysis. Model interpretability was achieved through Shapley Additive exPlanations (SHAP) analysis to quantify each biomarker's contribution. The clinically optimal model was deployed as a web-based graphical interface using Flask v2.3.2, enabling real-time probability estimation from six input parameters: albumin, γ-glutamyl transferase, total bilirubin, total bile acids, sex, and urea. 2.4 Statistical Analysis Framework Continuous variables with normal distributions are presented as mean ± standard deviation and analyzed using independent Student's t-tests. Non-normally distributed data were expressed as median (interquartile range) and compared via Mann-Whitney U tests. Categorical variables underwent evaluation through χ² tests or Fisher's exact tests with continuity correction where appropriate. Statistical significance was defined at two-sided P < 0.05. All analyses were performed using Python v3.10 (Python Software Foundation), R v4.2.1 (R Foundation for Statistical Computing), and SPSS v23.0 (IBM Corp.), with visualizations generated through Matplotlib v3.7.1 and SHAP v0.42.1 libraries. 3. Results 3.1 Baseline Demographic and Biochemical Characteristics This study enrolled 243 cholestatic neonates (age ≤ 30 days), comprising 61 with BA and 182 non-BA infants. Baseline age showed no significant difference between groups (18.18 ± 7.69 vs. 18.47 ± 7.63 days; P = 0.800). BA infants exhibited significantly more severe cholestatic profiles: γ-glutamyl transferase (GGT: 641.09 ± 367.30 U/L vs. 210.36 ± 183.84 U/L), total bilirubin (TBIL: 197.79 ± 78.41 µmol/L vs. 117.34 ± 80.17 µmol/L), direct bilirubin (DBIL: 91.07 ± 54.93 µmol/L vs. 52.80 ± 52.23 µmol/L), and total bile acids (TBA: 117.38 ± 75.04 µmol/L vs. 51.21 ± 44.56 µmol/L) were all markedly elevated (all P < 0.001). Paradoxically, BA infants demonstrated enhanced hepatic synthetic function evidenced by increased levels of albumin (ALB: 36.51 ± 3.19 g/L vs. 31.48 ± 6.80 g/L), total protein (TP: 52.78 ± 4.82 g/L vs. 48.49 ± 9.21 g/L), and albumin-to-globulin ratio (A/G: 2.33 ± 0.50 vs. 2.03 ± 0.85; all P < 0.001), despite reduced urea levels (3.19 ± 1.26 mmol/L vs. 6.30 ± 7.19 mmol/L; P < 0.001). Additionally, female predominance was significant in the BA group (63.93% vs. 34.62%; P < 0.001; Table 1 ). Table 1 Biochemical Data of the 243 BA and non-BA Neonatal Cohorts Variables Total (n = 243) Non-BA (n = 182) BA (n = 61) Statistic P Age (Mean ± SD, day) 18.40 ± 7.63 18.47 ± 7.63 18.18 ± 7.69 t = 0.25 0.800 Gender, n (%) χ² = 16.12 < 0.001 Female 102 (41.98) 63 (34.62) 39 (63.93) Male 141 (58.02) 119 (65.38) 22 (36.07) K (Mean ± SD, mmol/L) 4.83 ± 0.97 4.73 ± 1.04 5.11 ± 0.62 t = -2.70 0.007 Na (Mean ± SD, mmol/L) 135.71 ± 4.99 135.81 ± 5.61 135.42 ± 2.34 t = 0.76 0.450 Cl (Mean ± SD, mmol/L) 103.88 ± 5.67 103.70 ± 6.39 104.41 ± 2.45 t = -1.23 0.218 CO2 (Mean ± SD, mmol/L) 21.37 ± 4.93 21.67 ± 5.24 20.47 ± 3.72 t = 1.96 0.052 AG (Mean ± SD, mmol/L) 15.30 ± 5.09 15.17 ± 5.57 15.67 ± 3.24 t = -0.86 0.393 OSM (Mean ± SD, mOsm/kg) 281.57 ± 10.61 282.57 ± 11.88 278.59 ± 4.12 t = 3.88 < 0.001 TP (Mean ± SD, g/L) 49.57 ± 8.53 48.49 ± 9.21 52.78 ± 4.82 t = -4.66 < 0.001 PA (Mean ± SD, mg/L) 78.23 ± 40.18 76.32 ± 42.06 84.32 ± 33.07 t = -1.49 0.140 ALB (Mean ± SD, g/L) 32.75 ± 6.47 31.48 ± 6.80 36.51 ± 3.19 t = -7.75 < 0.001 GLO (Mean ± SD, g/L) 16.82 ± 4.80 17.01 ± 5.19 16.27 ± 3.38 t = 1.28 0.201 A/G (Mean ± SD) 2.11 ± 0.79 2.03 ± 0.85 2.33 ± 0.50 t = -3.36 < 0.001 Urea (Mean ± SD, mmol/L) 5.52 ± 6.40 6.30 ± 7.19 3.19 ± 1.26 t = 5.59 < 0.001 Cr (Mean ± SD, µmol/L) 37.72 ± 50.71 43.68 ± 57.26 19.93 ± 7.31 t = 5.46 < 0.001 Urea/Cr (Mean ± SD) 0.20 ± 0.26 0.20 ± 0.28 0.20 ± 0.18 t = 0.12 0.907 Chol (Mean ± SD, mmol/L) 2.98 ± 1.28 2.83 ± 1.26 3.47 ± 1.23 t = -3.35 < 0.001 UA (Mean ± SD, mmol/L) 184.21 ± 169.44 204.51 ± 189.04 123.62 ± 54.79 t = 5.16 < 0.001 GLU (Mean ± SD, mmol/L) 4.61 ± 1.61 4.62 ± 1.73 4.57 ± 1.21 t = 0.28 0.782 Ca (Mean ± SD, mmol/L) 2.32 ± 0.31 2.28 ± 0.34 2.47 ± 0.15 t = -6.11 < 0.001 P (Mean ± SD, mmol/L) 1.87 ± 0.63 1.82 ± 0.70 2.03 ± 0.29 t = -3.44 < 0.001 ALP (Mean ± SD, U/L) 359.47 ± 202.16 346.60 ± 213.50 397.89 ± 159.12 t = -1.72 0.086 AST (Mean ± SD, U/L) 97.19 ± 138.10 94.18 ± 150.51 106.16 ± 92.07 t = -0.59 0.559 ALT (Mean ± SD, U/L) 49.55 ± 73.36 47.57 ± 79.88 55.48 ± 49.08 t = -0.73 0.467 AST/ALT (Mean ± SD) 2.53 ± 2.20 2.63 ± 2.50 2.25 ± 0.81 t = 1.14 0.254 GGT (Mean ± SD, U/L) 318.49 ± 306.19 210.36 ± 183.84 641.09 ± 367.30 t = -8.80 < 0.001 TBIL (Mean ± SD, umol/L) 137.54 ± 86.91 117.34 ± 80.17 197.79 ± 78.41 t = -6.82 < 0.001 DBIL (Mean ± SD, umol/L) 62.41 ± 55.37 52.80 ± 52.23 91.07 ± 54.93 t = -4.89 < 0.001 IBIL (Mean ± SD, umol/L) 75.13 ± 56.13 64.54 ± 46.34 106.72 ± 69.80 t = -4.41 < 0.001 D/I (Mean ± SD) 1.35 ± 2.91 1.15 ± 1.43 1.97 ± 5.24 t = -1.22 0.228 TBA (Mean ± SD, umol/L) 67.82 ± 60.89 51.21 ± 44.56 117.38 ± 75.04 t = -6.51 < 0.001 Mg (Mean ± SD, mmol/L) 0.88 ± 0.15 0.86 ± 0.17 0.94 ± 0.08 t = -5.09 < 0.001 TG (Mean ± SD, mmol/L) 1.22 ± 1.47 1.29 ± 1.65 1.01 ± 0.48 t = 1.25 0.212 CK (Mean ± SD, U/L) 225.24 ± 512.95 243.76 ± 584.21 169.98 ± 165.86 t = 0.97 0.332 LDH (Mean ± SD, U/L) 521.27 ± 561.11 572.96 ± 634.45 367.05 ± 152.58 t = 4.04 < 0.001 HDL-C (Mean ± SD, mmol/L) 0.76 ± 0.39 0.75 ± 0.42 0.82 ± 0.26 t = -1.62 0.107 LDL-C (Mean ± SD, mmol/L) 1.55 ± 1.02 1.41 ± 0.96 2.00 ± 1.08 t = -3.94 < 0.001 VLDL-C (Mean ± SD, mmol/L) 0.40 ± 0.55 0.41 ± 0.58 0.37 ± 0.45 t = 0.51 0.607 Abbreviations: t, t-test; χ², Chi-square test; SD, standard deviation; P , P-value; K, potassium; Na, sodium; Cl, chloride; CO₂, carbon dioxide; AG, anion gap; OSM, osmolality; TP, total protein; PA, prealbumin; ALB, albumin; GLO, globulin; A/G, albumin/globulin ratio; Urea, urea; Cr, creatinine; Urea/Cr, urea/creatinine ratio; Chol, total cholesterol; UA, uric acid; GLU, glucose; Ca, calcium; P, phosphorus; ALP, alkaline phosphatase; AST, aspartate aminotransferase; ALT, alanine aminotransferase; AST/ALT, AST/ALT ratio; GGT, gamma-glutamyl transferase; TBIL, total bilirubin; DBIL, direct bilirubin; IBIL, indirect bilirubin; D/I, DBIL/ IBIL ratio; TBA, total bile acids; Mg, magnesium; TG, triglycerides; CK, creatine kinase; LDH, lactate dehydrogenase; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; VLDL-C, very-low-density lipoprotein cholesterol. 3.2 Univariate and Multivariate Predictors of Neonatal BA Univariable analysis identified 11 BA predictors ( P < 0.05), among which GGT (OR = 1.01, 95%CI: 1.01–1.01), TBIL (OR = 1.01, 95%CI: 1.01–1.01), and male sex (protective: OR = 0.30, 95%CI: 0.16–0.55) showed strongest associations. Multivariable analysis established five independent predictors: GGT (OR = 1.01, 95%CI: 1.01–1.01, P < 0.001), TBIL (OR = 1.01, 95%CI: 1.01–1.01, P = 0.010), albumin (ALB: OR = 1.17, 95%CI: 1.03–1.32, P = 0.014), total bile acids (TBA: OR = 1.02, 95%CI: 1.01–1.02, P < 0.001), and urea (Urea: OR = 0.75, 95%CI: 0.57–0.99, P = 0.041; Table 2 ). Table 2 ༎Univariate and Multivariate Analysis of Neonatal BA Diagnostic Characteristics Variables Univariate Analysis Multivariate Analysis β S.E Z P OR (95%CI) β S.E Z P OR (95%CI) Age -0.00 0.02 -0.25 0.799 1.00 (0.96–1.03) Gender Female 1.00 (Reference) Male -1.21 0.31 -3.91 < 0.001 0.30 (0.16–0.55) TP 0.07 0.02 3.30 < 0.001 1.07 (1.03–1.11) PA 0.00 0.00 1.31 0.191 1.00 (1.00-1.01) ALB 0.17 0.03 4.86 < 0.001 1.18 (1.11–1.27) 0.15 0.06 2.47 0.014 1.17 (1.03–1.32) Urea -0.34 0.09 -3.82 < 0.001 0.71 (0.60–0.85) -0.29 0.14 -2.04 0.041 0.75 (0.57–0.99) TBIL 0.01 0.00 5.65 < 0.001 1.01 (1.01–1.01) 0.01 0.00 2.56 0.010 1.01 (1.01–1.01) DBIL 0.01 0.00 4.17 < 0.001 1.01 (1.01–1.02) AST 0.00 0.00 0.58 0.560 1.00 (1.00–1.00) GGT 0.01 0.00 7.11 < 0.001 1.01 (1.01–1.01) 0.01 0.00 5.08 < 0.001 1.01 (1.01–1.01) ALP 0.00 0.00 1.68 0.093 1.00 (1.00–1.00) TBA 0.02 0.00 5.45 < 0.001 1.02 (1.02–1.03) 0.02 0.00 3.48 < 0.001 1.02 (1.01–1.02) TG -0.26 0.20 -1.26 0.207 0.77 (0.52–1.15) LDH -0.01 0.00 -2.40 0.016 0.99 (0.99–0.99) HDL-C 0.49 0.38 1.28 0.201 1.63 (0.77–3.45) LDL-C 0.55 0.15 3.60 < 0.001 1.73 (1.28–2.34) VLDL-C -0.15 0.30 -0.51 0.607 0.86 (0.48–1.54) Abbreviations: β, beta coefficient; S.E, standard error; Z, Z-value; P, P-value; OR (95%CI), odds ratio (95% confidence interval); TP, total protein; PA, prealbumin; ALB, albumin; Urea, urea; TBIL, total bilirubin; DBIL, direct bilirubin; AST, aspartate aminotransferase; GGT, gamma-glutamyl transferase; ALP, alkaline phosphatase; TBA, total bile acids; TG, triglycerides; LDH, lactate dehydrogenase; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; VLDL-C, very-low-density lipoprotein cholesterol. 3.3 Machine Learning-Based Diagnostic Screening Models for Neonatal BA Serum biomarkers from 243 neonates were used to construct five machine learning-based diagnostic screening models for early BA detection. Feature selection was validated by Spearman correlation (r < 0.9), with LASSO regression identifying six optimal predictors: ALB, GGT, TBIL, TBA, gender, and urea (Fig. 2 A and Supplementary Fig. 1). Following 10-fold cross-validation and Optuna hyperparameter tuning, all models demonstrated robust training performance (AUC > 0.8; Table 3 and Fig. 2 B). During training, support vector classifier (SVC; AUC = 0.962), and random forest (RF; AUC = 0.952) achieved the highest discriminative power. In the independent validation set, RF attained best classification (AUC = 0.915, 95%CI: 0.827-1.000; Fig. 2 C). RF exhibited exceptional reproducibility, with consistent AUCs in training (0.959) and validation (0.915) phases. Calibration curves confirmed high agreement between predicted and observed probabilities across all models (Fig. 2 D). Table 3 Diagnostic screening performance of five machine learning models for neonatal biliary atresia in the training and validation sets Algorithms Average performance of cross-validation on the training set Performance on the validation set Precision Recall F1-score Accuracy AUROC (95%CI) Precision Recall F1-score Accuracy AUROC (95%CI) LRM 0.867 0.455 0.578 0.853 0.937 (0.900-0.973) 1.000 0.500 0.667 0.863 0.750 (0.614–0.886) DET 0.708 0.685 0.678 0.853 0.796 (0.721–0.870) 0.621 0.900 0.735 0.822 0.846 (0.733–0.960) MLP 0.750 0.425 0.504 0.829 0.868 (0.815–0.921) 0.667 0.400 0.500 0.781 0.662 (0.516–0.809) SVC 0.855 0.705 0.743 0.894 0.962 (0.940–0.985) 0.750 0.750 0.750 0.890 0.828 (0.709–0.947) RF 0.843 0.805 0.815 0.912 0.952 (0.917–0.987) 0.690 1.000 0.816 0.877 0.915 (0.827-1.000) Abbreviation: AUROC, area under the receiver operating characteristic; CI, confidence interval; ML, machine learning; LRM, logistic regression model; RF, random forest; SVC, support vector machine classifier; MLP, multilayer perceptron; DET, decision tree 3.4 Clinical Deployment of the Neonatal BA Diagnostic Screening Platform Consistent feature importance analysis using SHAP summary plots (Fig. 3 A-F) established GGT, TBIL, and TBA as the most influential predictors across all five diagnostic screening models. SHAP dependence plots further elucidated biomarker thresholds and feature interactions, revealing monotonic relationships where elevated values (represented by red data points in Fig. 4 A-F) consistently increased BA probability. These computational insights drove the selection of the random forest model - distinguished by best validation performance (AUC = 0.915, Fig. 5 ) - for clinical implementation as a web-based diagnostic screening tool. The deployed interface (Fig. 5 A-B) features an intuitive input panel for six parameters: ALB (g/L), GGT (U/L), TBIL (µmol/L), TBA (µmol/L), sex (categorical), and urea (mmol/L). Upon submission, the platform generates comprehensive clinical outputs in < 2 seconds: (i) binary BA/non-BA classification, (ii) quantified risk probability (0-100%) stratified as low ( 80%), and (iii) an interpretable visualization of feature contributions modeled through SHAP force plots. Implementation resources including cross-platform installation packages are detailed in Supplementary Files. 4. Discussion In this study, we developed and validated a ML-based screening model using routine biochemical parameters and basic demographic data to assist in the early identification of BA among cholestatic neonates. The model exhibited excellent discriminative performance, with several algorithms, particularly RF and SVC, achieving near-perfect classification in the validation cohort. Furthermore, the model was successfully deployed as a web-based interface, enabling real-time clinical application at the bedside. GGT was identified as the most influential predictor across all machine learning models. Elevated GGT levels are known to be strongly associated with BA and reflect cholangiocyte proliferation and bile duct injury 19 . Prior studies have established GGT as a core diagnostic marker for BA, especially in distinguishing it from intrahepatic cholestasis 19 , 20 . However, previous research has also demonstrated a significant correlation between GGT levels and patient age in BA, suggesting that diagnostic models derived from older infants may not be directly applicable to neonates 21 . Our findings reaffirm the pivotal role of GGT in BA diagnosis and support its integration into ML-assisted diagnostic tools. TBIL and TBA also ranked among the top predictors in the machine learning models. Both are classical biochemical markers of cholestasis, but previous studies have shown that their levels tend to be more markedly elevated in BA compared to other causes of neonatal cholestasis. 9 , 22 . In our SHAP analysis, higher TBIL and TBA levels were associated with an increased predicted probability of BA, demonstrating a clear monotonic relationship. Although these indicators are routinely measured in clinical settings, their integration into ML models significantly improved diagnostic performance. These findings highlight the potential of combining conventional biochemical markers with advanced algorithms to enhance diagnostic accuracy in neonatal BA. Interestingly, ALB levels were significantly higher in the BA group compared to the non-BA group in our study. While hypoalbuminemia is typically associated with impaired hepatic function, this finding suggests that neonates with BA may be in an early disease stage, during which albumin synthesis remains preserved or even upregulated. This paradoxical elevation may reflect compensatory hepatic synthetic activity or early nutritional support. In contrast, non-BA cholestasis, such as parenteral nutrition-associated cholestasis, was often accompanied by acute hepatocellular injury, which can lead to early reductions in albumin production 23 . Incorporating ALB into our model enhanced its ability to capture early functional differences in the liver, supporting its utility as an auxiliary marker in neonatal BA screening. Serum urea levels were negatively associated with BA. Lower urea levels in BA infants may reflect reduced protein catabolism or early shifts in hepatic nitrogen metabolism. Female sex was also identified as an independent risk factor for BA, consistent with previous studies reporting a higher incidence of BA among female infants 24 , 25 . Choi et al. incorporated demographic variables such as age and sex, along with biochemical markers, into their diagnostic model for BA, achieving an AUC of 0.97 26 . Similarly, our study found that including sex as a variable improved the overall predictive performance of the model, highlighting the value of integrating demographic data into early diagnostic strategies. Previous ML-based studies on BA have frequently incorporated advanced imaging features, liver stiffness measurements, or novel biomarkers such as MMP-7 5, 15, 27 . Among these, MMP-7 has emerged as a promising serological marker with high diagnostic accuracy, showing AUC values consistently above 0.90 in multiple studies 18 , 28 . However, several limitations hinder its integration into routine clinical practice, including high testing cost, limited accessibility in many hospitals, and lack of standardized cutoff values across age groups. In contrast, our ML model is built entirely upon routine biochemical parameters that are universally available, inexpensive, and already part of standard neonatal cholestasis workups. Therefore, the simplicity, reproducibility, and affordability of our model position it as a practical alternative for early BA screening, especially in resource-constrained regions. The primary strengths of our study include its data-driven feature selection, rigorous model validation, and clinically deployable web interface. However, this work has some limitations. First, it was based on a single-center retrospective cohort, which may limit generalizability. External validation using multicenter datasets is needed. Second, the model excluded imaging features that may offer additional diagnostic value. Third, real-world clinical impact of the online tool requires prospective testing in diverse practice settings. 5. Conclusion In conclusion, this study developed diagnostic screening models for neonatal biliary atresia using routine biochemical markers and demographic characteristics. The random forest model showed the highest predictive accuracy and may serve as a practical screening tool to aid in the early identification of BA. Although further external validation is required, these findings suggest that machine learning-based diagnostic screening models hold promise for improving the diagnostic pathway and clinical management of neonatal BA. Declarations 6.1 Acknowledgments The authors would like to thank all the medical and nursing staff from the Department of Neonatal Surgery, Beijing Children’s Hospital, Capital Medical University, for their dedicated patient care and assistance in data collection. We are also grateful to the Central Laboratory of Jiangxi Provincial Children’s Hospital for providing technical support during this study. 6.2 Funding This work was supported by the Beijing Municipal Natural Science Foundation (Grant No. 7252043 to JSH), the Jiangxi (Ganpo) Talent Program-Health Innovation Talent Project (to JSH), and the National Natural Science Foundation of China (Grant No. 82400592 to JDW; Grant No. 82300574 to DYS). 6.3 Conflict of Interest The authors declare no conflicts of interest related to this publication. 6.4 Author Contributions Study conception and design (JSH, DDW); data acquisition (ZZL, YYJ, YZ, YNZ, SSL, JML, KYH, YCG, DYS); data analysis and interpretation (ZZL, YYJ, JSH, DDW); drafting of the manuscript (ZZL, YYJ); critical revision of the manuscript for important intellectual content (JSH, JDW). All authors have read and approved the final manuscript. Zhaozhou Liu and Yuyan Jin contributed equally as co-first authors. 6.5 Ethical Statement This study was approved by the Ethics Committee of Beijing Children’s Hospital, Capital Medical University (Approval No. 2019-k-386). The requirement for informed consent was waived due to the retrospective and anonymized design of the study. The research was conducted in accordance with the principles of the Declaration of Helsinki. 6.6 Data Sharing Statement The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request. No additional data are available. References Chung PHY, Zheng S, Tam PKH. Biliary atresia: East versus west. Seminars in pediatric surgery. 2020;29:150950. Tam PKH, Wells RG, Tang CSM, et al. Biliary atresia. Nature reviews. Disease primers. 2024;10:47. Fanna M, Masson G, Capito C, et al. Management of Biliary Atresia in France 1986 to 2015: Long-term Results. Journal of pediatric gastroenterology and nutrition. 2019;69:416-424. Xie C, Wang P, Wang D, et al. Is performing the Kasai portoenterostomy in the neonatal period associated with a better prognosis? A single-center, retrospective cohort study from China. BMC pediatrics. 2025;25:454. Hou J, Xiao W, Zhou S, Liu H. Identification of Biliary Atresia in Infantile Cholestasis: Integrating Radiomics With MRCP for Unobservable Extrahepatic Biliary Systems. Journal of computer assisted tomography. 2025. Kianifar HR, Tehranian S, Shojaei P, et al. Accuracy of hepatobiliary scintigraphy for differentiation of neonatal hepatitis from biliary atresia: systematic review and meta-analysis of the literature. Pediatric radiology. 2013;43:905-919. Ma Y, Yang Y, Du Y, et al. Development of an artificial intelligence-based multimodal diagnostic system for early detection of biliary atresia. BMC medicine. 2025;23:127. Brahee DD, Lampl BS. Neonatal diagnosis of biliary atresia: a practical review and update. Pediatric radiology. 2022;52:685-692. Harpavat S, Garcia-Prats JA, Anaya C, et al. Diagnostic Yield of Newborn Screening for Biliary Atresia Using Direct or Conjugated Bilirubin Measurements. JAMA. 2020;323:1141-1150. Gong Z, Lin L, Lu G, Wan C. Development and validation of a model for early diagnosis of biliary atresia. BMC pediatrics. 2023;23:549. Golden J, Zagory JA, Fenlon M, et al. Liquid chromatography-mass spectroscopy in the diagnosis of biliary atresia in children with hyperbilirubinemia. The Journal of surgical research. 2018;228:228-237. Pandurangi S, Mourya R, Nalluri S, et al. Diagnostic accuracy of serum matrix metalloproteinase-7 as a biomarker of biliary atresia in a large North American cohort. Hepatology (Baltimore, Md.). 2024;80:152-162. Fu M, Guo Z, Chen Y, et al. Proteomics Defines Plasma Biomarkers for the Early Diagnosis of Biliary Atresia. Journal of proteome research. 2024;23:1744-1756. Lyu H, Ye Y, Lui VCH, et al. Plasma amyloid-beta levels correlated with impaired hepatic functions: An adjuvant biomarker for the diagnosis of biliary atresia. Frontiers in surgery. 2022;9:931637. Zhou W, Yang Y, Yu C, et al. Ensembled deep learning model outperforms human experts in diagnosing biliary atresia from sonographic gallbladder images. Nature communications. 2021;12:1259. Hsu F-R, Dai S-T, Chou C-M, Huang S-Y. The application of artificial intelligence to support biliary atresia screening by ultrasound images: A study based on deep learning models. PloS one. 2022;17:e0276278. Sun Y, Dai S, Shen Z, et al. Gamma-glutamyl transpeptidase has different efficacy on biliary atresia diagnosis in different hospital patient groups: an application of machine learning approach. Pediatric surgery international. 2022;38:1131-1141. Zhao Y, Wang A, Wang D, et al. Development of a diagnostic model for biliary atresia based on MMP7 and serological tests using machine learning. Pediatric surgery international. 2024;40:203. Kong F, Dong R, Chen G, et al. Progress in Biomarkers Related to Biliary Atresia. Journal of clinical and translational hepatology. 2024;12:305-315. Hsiao Y-H, Hung W-L, Yang Y-J, et al. Non-invasive simple predictors of biliary atresia in cholestatic infants - A preliminary report. Pediatrics and neonatology. 2025. Chen X, Dong R, Shen Z, Yan W, Zheng S. Value of Gamma-Glutamyl Transpeptidase for Diagnosis of Biliary Atresia by Correlation With Age. Journal of pediatric gastroenterology and nutrition. 2016;63:370-373. Han Y-J, Hu S-Q, Zhu J-H, et al. Accurate prediction of biliary atresia with an integrated model using MMP-7 levels and bile acids. World journal of pediatrics : WJP. 2024;20:822-833. Feldman AG, Sokol RJ. Neonatal Cholestasis: Updates on Diagnostics, Therapeutics, and Prevention. NeoReviews. 2021;22:e819-e836. Lee KJ, Kim JW, Moon JS, Ko JS. Epidemiology of Biliary Atresia in Korea. Journal of Korean medical science. 2017;32:656-660. Hukkinen M, Kerola A, Lohi J, et al. Treatment Policy and Liver Histopathology Predict Biliary Atresia Outcomes: Results after National Centralization and Protocol Biopsies. Journal of the American College of Surgeons. 2018;226:46-57.e41. Choi HJ, Kim YE, Namgoong J-M, et al. Development and Validation of a Machine Learning-Based Prediction Model for Detection of Biliary Atresia. Gastro hep advances. 2023;2:778-787. Chen X, Zhao D, Ji H, Chen Y, Li Y, Zuo Z. Predictive modeling for early detection of biliary atresia in infants with cholestasis: Insights from a machine learning study. Computers in biology and medicine. 2024;174:108439. Zhang Y, Li T, Wang T, Ji Q, Zhan J. Comparison for the diagnostic performance of early diagnostic methods for biliary atresia: a systematic review and network meta-analysis. Pediatric surgery international. 2024;40:146. Additional Declarations No competing interests reported. Supplementary Files SupplementaryFigure1.jpg Cite Share Download PDF Status: Published Journal Publication published 20 Dec, 2025 Read the published version in Pediatric Surgery International → Version 1 posted Editorial decision: Revision requested 16 Oct, 2025 Reviews received at journal 13 Oct, 2025 Reviewers agreed at journal 22 Sep, 2025 Reviewers invited by journal 02 Sep, 2025 Editor assigned by journal 29 Aug, 2025 Submission checks completed at journal 28 Aug, 2025 First submitted to journal 26 Aug, 2025 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-7460903","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":510244060,"identity":"062735b1-f466-46ee-8ad0-acc6d2520d05","order_by":0,"name":"Zhaozhou Liu","email":"","orcid":"","institution":"Capital Medical University, National Center for Children's Health","correspondingAuthor":false,"prefix":"","firstName":"Zhaozhou","middleName":"","lastName":"Liu","suffix":""},{"id":510244061,"identity":"0d7bb634-6949-47d5-a48e-fc74314dbe5a","order_by":1,"name":"Yuyan Jin","email":"","orcid":"","institution":"Capital Medical University, National Center for Children's Health","correspondingAuthor":false,"prefix":"","firstName":"Yuyan","middleName":"","lastName":"Jin","suffix":""},{"id":510244062,"identity":"181377c9-41c0-4d97-a488-01f6753c4379","order_by":2,"name":"Yong Zhao","email":"","orcid":"","institution":"Capital Medical University, National Center for Children's Health","correspondingAuthor":false,"prefix":"","firstName":"Yong","middleName":"","lastName":"Zhao","suffix":""},{"id":510244063,"identity":"02499a0b-06d7-4ab8-8194-c11f291fdb2f","order_by":3,"name":"Yanan Zhang","email":"","orcid":"","institution":"Capital Medical University, National Center for Children's Health","correspondingAuthor":false,"prefix":"","firstName":"Yanan","middleName":"","lastName":"Zhang","suffix":""},{"id":510244064,"identity":"748ceddf-fbd9-418f-8c75-19cdbacf7873","order_by":4,"name":"Shuangshuang Li","email":"","orcid":"","institution":"Capital Medical University, National Center for Children's Health","correspondingAuthor":false,"prefix":"","firstName":"Shuangshuang","middleName":"","lastName":"Li","suffix":""},{"id":510244066,"identity":"a4b2c972-2c23-4707-a358-2ce1f6b1de49","order_by":5,"name":"Junmin Liao","email":"","orcid":"","institution":"Capital Medical University, National Center for Children's Health","correspondingAuthor":false,"prefix":"","firstName":"Junmin","middleName":"","lastName":"Liao","suffix":""},{"id":510244067,"identity":"8b97e131-3cd8-4ef0-b675-de590b52dde4","order_by":6,"name":"Kaiyun Hua","email":"","orcid":"","institution":"Capital Medical University, National Center for Children's Health","correspondingAuthor":false,"prefix":"","firstName":"Kaiyun","middleName":"","lastName":"Hua","suffix":""},{"id":510244069,"identity":"968d092e-2889-4451-aca4-c6deac931a45","order_by":7,"name":"Yichao Gu","email":"","orcid":"","institution":"Capital Medical University, National Center for Children's Health","correspondingAuthor":false,"prefix":"","firstName":"Yichao","middleName":"","lastName":"Gu","suffix":""},{"id":510244073,"identity":"acef922e-2e91-488c-a3a5-0442492034a9","order_by":8,"name":"Dayan Sun","email":"","orcid":"","institution":"Capital Medical University, National Center for Children's Health","correspondingAuthor":false,"prefix":"","firstName":"Dayan","middleName":"","lastName":"Sun","suffix":""},{"id":510244074,"identity":"428adad2-a87b-494a-b908-b8aa9f239046","order_by":9,"name":"Dingding Wang","email":"","orcid":"","institution":"Capital Medical University, National Center for Children's Health","correspondingAuthor":false,"prefix":"","firstName":"Dingding","middleName":"","lastName":"Wang","suffix":""},{"id":510244075,"identity":"9be411d4-4b7e-4f9d-af35-b997388c255d","order_by":10,"name":"Jinshi Huang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAArElEQVRIiWNgGAWjYBACA4YDbAyMDTY8/PwNpGlJk5GccYBoLQwgLYdtDBoSiNRiznj82YOfO87zAK1j/PAxhwgtlg1nzA17z9zmMWduYJacuY0Yhx04wybN2Habx7LhABszL3Fajj8DajnHY3AggWgtB8yAWg6QpAXkl7ZkHskZB5uJ9MsNUIi12dnz8zcf/PCRGC0MEgdgLMYGYtQDAdHpZBSMglEwCkYuAADxeTsZyXSwngAAAABJRU5ErkJggg==","orcid":"","institution":"Capital Medical University, National Center for Children's Health","correspondingAuthor":true,"prefix":"","firstName":"Jinshi","middleName":"","lastName":"Huang","suffix":""}],"badges":[],"createdAt":"2025-08-26 08:53:47","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7460903/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7460903/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00383-025-06258-6","type":"published","date":"2025-12-20T15:57:44+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":90899208,"identity":"44cb1e33-fce6-4fe2-9e84-e66df691d980","added_by":"auto","created_at":"2025-09-09 11:58:38","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":94389,"visible":true,"origin":"","legend":"\u003cp\u003eStudy flowchart of neonatal cholestasis cohort. Patient enrollment diagram for 243 jaundiced neonates (≤30 days) evaluated at Beijing Children's Hospital (2017-2023). Inclusion required biochemical cholestasis criteria (DBIL \u0026gt;17.1 μmol/L when TBIL ≤85.5 μmol/L, or DBIL \u0026gt;20% of TBIL when TBIL \u0026gt;85.5 μmol/L). BA diagnosis (n=61) was confirmed by intraoperative cholangiography; non-BA cholestasis (n=182) was determined through multimodal assessment (cholangiography, biopsy, genetic testing, and clinical follow-up). Exclusions comprised cases with incomplete workup or indeterminate etiology. The cohort was stratified by diagnosis and randomly partitioned (7:3 ratio) for model training and validation. Abbreviations: BA, biliary atresia; BCH, Beijing Children's Hospital; DBIL, direct bilirubin; IOC, intraoperative cholangiography; TBIL, total bilirubin; ML, machine learning.\u003c/p\u003e","description":"","filename":"Figure1.tif.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7460903/v1/85f4d4e02f23f5df24f4d926.jpg"},{"id":90899209,"identity":"d7cb236b-3538-4f62-b2d3-800880b6fdba","added_by":"auto","created_at":"2025-09-09 11:58:38","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":4302822,"visible":true,"origin":"","legend":"\u003cp\u003eDevelopment and performance evaluation of machine learning neonatal BA diagnostic models. (A) Feature correlation matrix after LASSO selection (ALB, GGT, TBIL, TBA, gender, urea) demonstrating absence of strong collinearity (|r| \u0026lt; 0.9). (B) Receiver operating characteristic (ROC) curves for five ML models on the training cohort (n=170). (C) Validation cohort ROC curves (n=73): RF achieved best discrimination (AUC=0.915, 95% CI: 0.827-1.000). (D) Calibration curves showing agreement between predicted probabilities and observed frequencies; dashed line represents ideal calibration. Abbreviations: ALB, albumin; AUC, area under ROC curve; CI, confidence interval; DET, decision tree; GGT, γ-glutamyltransferase; LASSO, least absolute shrinkage and selection operator; LR, logistic regression; MLP, multilayer perceptron; RF, random forest; SVC, support vector classifier.\u003c/p\u003e","description":"","filename":"Figure2.tif.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7460903/v1/1740c83e1a9cb5e38db9818e.jpg"},{"id":90899845,"identity":"94340247-2854-4077-b43f-93cd707c9c4d","added_by":"auto","created_at":"2025-09-09 12:06:37","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":2147110,"visible":true,"origin":"","legend":"\u003cp\u003eFeature importance for neonatal BA diagnosis across algorithms. SHAP summary plots quantifying predictive contributions of key features in: (A) Logistic regression, (B) Decision tree, (C) Multilayer perceptron, (D) Support vector classifier, and (E) Random forest. GGT, TBIL, and TBA consistently emerged as the strongest BA predictors across all algorithms (higher SHAP values indicate greater diagnostic contribution). Abbreviations: BA, biliary atresia; SHAP, Shapley Additive exPlanations.\u003c/p\u003e","description":"","filename":"Figure3.tif.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7460903/v1/27fbfc045f0457fa8f4f0ea5.jpg"},{"id":90900975,"identity":"a003c9a3-6b32-4f7b-965e-127eb4b05038","added_by":"auto","created_at":"2025-09-09 12:14:37","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2233918,"visible":true,"origin":"","legend":"\u003cp\u003eDirectional effects of feature values on neonatal BA prediction. SHAP dependence plots illustrating how biomarker values modulate BA probability predictions in: (A) Logistic regression, (B) Decision tree, (C) Multilayer perceptron, (D) Support vector classifier, and (E) Random forest. Individual neonates are represented as points (red: high feature values; blue: low values). The vertical spread reflects interaction effects with other features. Abbreviations: BA, biliary atresia; SHAP, Shapley Additive exPlanations.\u003c/p\u003e","description":"","filename":"Figure4.tif.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7460903/v1/50eec49998b94342d2762f78.jpg"},{"id":90899846,"identity":"d24734b8-c211-4f9f-9396-534ed4bbd380","added_by":"auto","created_at":"2025-09-09 12:06:37","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":787902,"visible":true,"origin":"","legend":"\u003cp\u003eClinical deployment of the neonatal-specific biliary atresia screening tool. (A) Web-based graphical interface for point-of-care input of six neonatal parameters (ALB, GGT, TBIL, TBA, gender, urea) in infants \u0026lt;28 days. (B) Screening output display providing: (i) risk classification (BA/non-BA), (ii) predicted probability, and (iii) feature contribution weights. This real-time bedside screening tool employs the optimized random forest model to facilitate early detection of biliary atresia during the critical neonatal window. Abbreviations: ALB, albumin; BA, biliary atresia; GGT, γ-glutamyltransferase; TBA, total bile acid; TBIL, total bilirubin.\u003c/p\u003e","description":"","filename":"Figure5.tif.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7460903/v1/262ebe74a0048ed4fbfd243d.jpg"},{"id":102298283,"identity":"ede9bb3e-4407-4f5f-8ab0-efbc71b1c40d","added_by":"auto","created_at":"2026-02-10 10:36:39","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":11291690,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7460903/v1/975cd62e-d46b-44f6-befb-200183f4352b.pdf"},{"id":90899186,"identity":"0eb7d8b8-b484-439c-a04a-4a831716a602","added_by":"auto","created_at":"2025-09-09 11:58:37","extension":"jpg","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":2188398,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7460903/v1/d5f42abda82eb5b749825f26.jpg"}],"financialInterests":"No competing interests reported.","formattedTitle":"Development and Validation of a Screening Model for Early Diagnosis of Biliary Atresia in Neonates with Cholestasis","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eBiliary atresia (BA) is a severe neonatal cholestatic liver disease that causes progressive fibrosis and obstruction of both intrahepatic and extrahepatic bile ducts. The disease affects approximately 1 in 5,000 to 20,000 live births, with the highest incidence reported in Asian populations \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Without treatment, BA rapidly progresses to cirrhosis and liver failure, and it represents one of the leading indications for liver transplantation in infants \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Kasai portoenterostomy is the standard surgical treatment for BA. Early diagnosis followed by timely Kasai surgery is essential for improving outcomes because it helps preserve native liver function and may delay or even prevent liver transplantation \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. Our previous work demonstrated that infants who undergo Kasai surgery during the neonatal period achieve higher rates of postoperative jaundice clearance and native liver survival \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. However, early diagnosis of BA remains challenging in clinical practice.\u003c/p\u003e\u003cp\u003eClinical manifestations such as persistent jaundice and acholic stools often appear early but are nonspecific and overlap with other causes of neonatal cholestasis. Diagnostic imaging techniques, including abdominal ultrasound, hepatobiliary scintigraphy, and magnetic resonance cholangiopancreatography, are widely used, yet their performance depends heavily on operator expertise and disease stage \u003csup\u003e\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Liver biopsy may provide histological support but is invasive and may lack specificity in early-stage disease \u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. To improve early recognition, recent studies have evaluated serum biochemical markers. Conventional parameters such as gamma-glutamyl transferase (GGT), total bilirubin, direct bilirubin, and total bile acids are helpful in differentiating BA from non-BA cholestasis \u003csup\u003e\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. In addition, novel biomarkers such as matrix metalloproteinase-7 (MMP-7) and amyloid-beta show promising diagnostic value \u003csup\u003e\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. However, their widespread clinical use remains restricted because of high cost, technical complexity, and limited availability. By contrast, routine biochemical tests are inexpensive, universally available, and already integrated into neonatal care, making them an ideal basis for developing practical diagnostic screening models.\u003c/p\u003e\u003cp\u003eMachine learning (ML) techniques have gained increasing importance in medical diagnostics. These methods can detect complex data patterns and identify nonlinear relationships that conventional approaches often miss. In the context of BA, ML-based models have integrated symptoms, laboratory data, and imaging findings to improve diagnostic accuracy \u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan additionalcitationids=\"CR16 CR17\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. However, most existing studies focus on mixed-age pediatric populations, and few specifically address ML-assisted early diagnosis of neonatal BA. Moreover, most models have not yet been adapted for practical use at the bedside.\u003c/p\u003e\u003cp\u003eTherefore, the present study aimed to develop and validate a machine learning-based diagnostic screening model for neonatal BA using routine serum biochemical indicators and basic demographic data. Furthermore, we deployed the optimal model as a web-based interactive platform to enable rapid, bedside clinical application and to facilitate early identification of infants at risk for BA.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Patient Selection and Data Collection\u003c/h2\u003e\u003cp\u003eThis retrospective cohort study included cholestatic neonates evaluated at Beijing Children\u0026rsquo;s Hospital, Capital Medical University, between January 2017 and December 2023. Cholestasis was defined according to established biochemical criteria: (1) direct/conjugated bilirubin\u0026thinsp;\u0026gt;\u0026thinsp;17.1 \u0026micro;mol/L when total bilirubin (TBIL)\u0026thinsp;\u0026le;\u0026thinsp;85.5 \u0026micro;mol/L, or (2) direct/conjugated bilirubin fraction\u0026thinsp;\u0026gt;\u0026thinsp;20% of TBIL when TBIL\u0026thinsp;\u0026gt;\u0026thinsp;85.5 \u0026micro;mol/L. The diagnosis of BA was confirmed by intraoperative cholangiography. Non-BA etiologies were determined using a combination of intraoperative cholangiography, liver histopathology, genetic testing, and clinical response to conservative treatment. Eligible participants were neonates aged\u0026thinsp;\u0026le;\u0026thinsp;30 days who presented with biochemically confirmed cholestasis. Exclusion criteria included incomplete diagnostic records, indeterminate final diagnosis, or lack of follow-up at our institution. Ethical approval was obtained from the Institutional Review Board of Beijing Children\u0026rsquo;s Hospital (Approval No. 2019-k-386), with informed consent waived because of the retrospective and anonymized study design.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Data Collection and Biochemical Parameters\u003c/h2\u003e\u003cp\u003eComprehensive demographic characteristics and baseline laboratory parameters were extracted from the electronic medical records during the first clinical encounter prior to therapeutic intervention. The complete biochemical panel comprised 37 parameters across three physiological domains: (1) Electrolytes and renal function markers including potassium (K), sodium (Na), chloride (Cl), carbon dioxide (CO₂), anion gap, osmolality, creatinine, urea, and urea-to-creatinine ratio; (2) Hepatic metabolic markers encompassing total protein, albumin, globulin, albumin-to-globulin ratio, total cholesterol, triglycerides, glucose, calcium, phosphorus, magnesium, uric acid, and lipid fractions (HDL-C, LDL-C, VLDL-C); and (3) Cholestasis-specific biomarkers such as total bilirubin, direct bilirubin, indirect bilirubin, total bile acids, alkaline phosphatase, γ-glutamyl transferase, aspartate aminotransferase, alanine aminotransferase, AST-to-ALT ratio, prealbumin, creatine kinase, and lactate dehydrogenase.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Development and Validation of Machine Learning Algorithms\u003c/h2\u003e\u003cp\u003eWe applied five supervised machine learning algorithms to construct diagnostic screening models for neonatal BA. These algorithms included logistic regression model (LRM), decision tree (DET), multilayer perceptron (MLP), support vector machine classifier (SVC), and random forest (RF). All analyses were performed in Python 3.10 using the scikit-learn (v1.2.2) library. The cohort of 243 neonates with cholestasis (61 BA, 182 non-BA) was randomly partitioned into a training set (70%, n\u0026thinsp;=\u0026thinsp;170) and a validation set (30%, n\u0026thinsp;=\u0026thinsp;73) using stratified sampling to balance age and sex distributions. Feature engineering comprised multicollinearity reduction through exclusion of highly correlated variables (|Spearman's r| \u0026gt;0.90), followed by feature selection using L1-penalized LASSO regression. Each model underwent rigorous optimization through 10-fold cross-validation and hyperparameter tuning via Optuna framework. Performance was evaluated in validation cohorts using six core metrics: area under receiver operating characteristic curve (AUC-ROC), accuracy, precision, recall, F1-score, and calibration curve analysis. Model interpretability was achieved through Shapley Additive exPlanations (SHAP) analysis to quantify each biomarker's contribution. The clinically optimal model was deployed as a web-based graphical interface using Flask v2.3.2, enabling real-time probability estimation from six input parameters: albumin, γ-glutamyl transferase, total bilirubin, total bile acids, sex, and urea.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Statistical Analysis Framework\u003c/h2\u003e\u003cp\u003eContinuous variables with normal distributions are presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation and analyzed using independent Student's t-tests. Non-normally distributed data were expressed as median (interquartile range) and compared via Mann-Whitney U tests. Categorical variables underwent evaluation through χ\u0026sup2; tests or Fisher's exact tests with continuity correction where appropriate. Statistical significance was defined at two-sided \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05. All analyses were performed using Python v3.10 (Python Software Foundation), R v4.2.1 (R Foundation for Statistical Computing), and SPSS v23.0 (IBM Corp.), with visualizations generated through Matplotlib v3.7.1 and SHAP v0.42.1 libraries.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e3.1 Baseline Demographic and Biochemical Characteristics\u003c/h2\u003e\u003cp\u003eThis study enrolled 243 cholestatic neonates (age\u0026thinsp;\u0026le;\u0026thinsp;30 days), comprising 61 with BA and 182 non-BA infants. Baseline age showed no significant difference between groups (18.18\u0026thinsp;\u0026plusmn;\u0026thinsp;7.69 vs. 18.47\u0026thinsp;\u0026plusmn;\u0026thinsp;7.63 days; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.800). BA infants exhibited significantly more severe cholestatic profiles: γ-glutamyl transferase (GGT: 641.09\u0026thinsp;\u0026plusmn;\u0026thinsp;367.30 U/L vs. 210.36\u0026thinsp;\u0026plusmn;\u0026thinsp;183.84 U/L), total bilirubin (TBIL: 197.79\u0026thinsp;\u0026plusmn;\u0026thinsp;78.41 \u0026micro;mol/L vs. 117.34\u0026thinsp;\u0026plusmn;\u0026thinsp;80.17 \u0026micro;mol/L), direct bilirubin (DBIL: 91.07\u0026thinsp;\u0026plusmn;\u0026thinsp;54.93 \u0026micro;mol/L vs. 52.80\u0026thinsp;\u0026plusmn;\u0026thinsp;52.23 \u0026micro;mol/L), and total bile acids (TBA: 117.38\u0026thinsp;\u0026plusmn;\u0026thinsp;75.04 \u0026micro;mol/L vs. 51.21\u0026thinsp;\u0026plusmn;\u0026thinsp;44.56 \u0026micro;mol/L) were all markedly elevated (all \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Paradoxically, BA infants demonstrated enhanced hepatic synthetic function evidenced by increased levels of albumin (ALB: 36.51\u0026thinsp;\u0026plusmn;\u0026thinsp;3.19 g/L vs. 31.48\u0026thinsp;\u0026plusmn;\u0026thinsp;6.80 g/L), total protein (TP: 52.78\u0026thinsp;\u0026plusmn;\u0026thinsp;4.82 g/L vs. 48.49\u0026thinsp;\u0026plusmn;\u0026thinsp;9.21 g/L), and albumin-to-globulin ratio (A/G: 2.33\u0026thinsp;\u0026plusmn;\u0026thinsp;0.50 vs. 2.03\u0026thinsp;\u0026plusmn;\u0026thinsp;0.85; all \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), despite reduced urea levels (3.19\u0026thinsp;\u0026plusmn;\u0026thinsp;1.26 mmol/L vs. 6.30\u0026thinsp;\u0026plusmn;\u0026thinsp;7.19 mmol/L; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Additionally, female predominance was significant in the BA group (63.93% vs. 34.62%; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eBiochemical Data of the 243 BA and non-BA Neonatal Cohorts\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003cp\u003e(n\u0026thinsp;=\u0026thinsp;243)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eNon-BA\u003c/p\u003e\u003cp\u003e(n\u0026thinsp;=\u0026thinsp;182)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eBA\u003c/p\u003e\u003cp\u003e(n\u0026thinsp;=\u0026thinsp;61)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eStatistic\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, day)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e18.40\u0026thinsp;\u0026plusmn;\u0026thinsp;7.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e18.47\u0026thinsp;\u0026plusmn;\u0026thinsp;7.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e18.18\u0026thinsp;\u0026plusmn;\u0026thinsp;7.69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;0.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.800\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGender, n (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eχ\u0026sup2; = 16.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e102 (41.98)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e63 (34.62)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e39 (63.93)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e141 (58.02)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e119 (65.38)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e22 (36.07)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eK (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4.83\u0026thinsp;\u0026plusmn;\u0026thinsp;0.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4.73\u0026thinsp;\u0026plusmn;\u0026thinsp;1.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e5.11\u0026thinsp;\u0026plusmn;\u0026thinsp;0.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -2.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.007\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNa (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e135.71\u0026thinsp;\u0026plusmn;\u0026thinsp;4.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e135.81\u0026thinsp;\u0026plusmn;\u0026thinsp;5.61\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e135.42\u0026thinsp;\u0026plusmn;\u0026thinsp;2.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;0.76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.450\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCl (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e103.88\u0026thinsp;\u0026plusmn;\u0026thinsp;5.67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e103.70\u0026thinsp;\u0026plusmn;\u0026thinsp;6.39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e104.41\u0026thinsp;\u0026plusmn;\u0026thinsp;2.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -1.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.218\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCO2 (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e21.37\u0026thinsp;\u0026plusmn;\u0026thinsp;4.93\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e21.67\u0026thinsp;\u0026plusmn;\u0026thinsp;5.24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e20.47\u0026thinsp;\u0026plusmn;\u0026thinsp;3.72\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;1.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.052\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAG (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e15.30\u0026thinsp;\u0026plusmn;\u0026thinsp;5.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e15.17\u0026thinsp;\u0026plusmn;\u0026thinsp;5.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e15.67\u0026thinsp;\u0026plusmn;\u0026thinsp;3.24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -0.86\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.393\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOSM (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mOsm/kg)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e281.57\u0026thinsp;\u0026plusmn;\u0026thinsp;10.61\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e282.57\u0026thinsp;\u0026plusmn;\u0026thinsp;11.88\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e278.59\u0026thinsp;\u0026plusmn;\u0026thinsp;4.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;3.88\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTP (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, g/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e49.57\u0026thinsp;\u0026plusmn;\u0026thinsp;8.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e48.49\u0026thinsp;\u0026plusmn;\u0026thinsp;9.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e52.78\u0026thinsp;\u0026plusmn;\u0026thinsp;4.82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -4.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePA (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mg/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e78.23\u0026thinsp;\u0026plusmn;\u0026thinsp;40.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e76.32\u0026thinsp;\u0026plusmn;\u0026thinsp;42.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e84.32\u0026thinsp;\u0026plusmn;\u0026thinsp;33.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -1.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.140\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eALB (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, g/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e32.75\u0026thinsp;\u0026plusmn;\u0026thinsp;6.47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e31.48\u0026thinsp;\u0026plusmn;\u0026thinsp;6.80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e36.51\u0026thinsp;\u0026plusmn;\u0026thinsp;3.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -7.75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGLO (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, g/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e16.82\u0026thinsp;\u0026plusmn;\u0026thinsp;4.80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e17.01\u0026thinsp;\u0026plusmn;\u0026thinsp;5.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e16.27\u0026thinsp;\u0026plusmn;\u0026thinsp;3.38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;1.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.201\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eA/G (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.11\u0026thinsp;\u0026plusmn;\u0026thinsp;0.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.03\u0026thinsp;\u0026plusmn;\u0026thinsp;0.85\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.33\u0026thinsp;\u0026plusmn;\u0026thinsp;0.50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -3.36\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUrea (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e5.52\u0026thinsp;\u0026plusmn;\u0026thinsp;6.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e6.30\u0026thinsp;\u0026plusmn;\u0026thinsp;7.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3.19\u0026thinsp;\u0026plusmn;\u0026thinsp;1.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;5.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCr (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, \u0026micro;mol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e37.72\u0026thinsp;\u0026plusmn;\u0026thinsp;50.71\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e43.68\u0026thinsp;\u0026plusmn;\u0026thinsp;57.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e19.93\u0026thinsp;\u0026plusmn;\u0026thinsp;7.31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;5.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUrea/Cr (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.20\u0026thinsp;\u0026plusmn;\u0026thinsp;0.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.20\u0026thinsp;\u0026plusmn;\u0026thinsp;0.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.20\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;0.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.907\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eChol (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.98\u0026thinsp;\u0026plusmn;\u0026thinsp;1.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.83\u0026thinsp;\u0026plusmn;\u0026thinsp;1.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3.47\u0026thinsp;\u0026plusmn;\u0026thinsp;1.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -3.35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUA (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e184.21\u0026thinsp;\u0026plusmn;\u0026thinsp;169.44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e204.51\u0026thinsp;\u0026plusmn;\u0026thinsp;189.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e123.62\u0026thinsp;\u0026plusmn;\u0026thinsp;54.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;5.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGLU (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4.61\u0026thinsp;\u0026plusmn;\u0026thinsp;1.61\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4.62\u0026thinsp;\u0026plusmn;\u0026thinsp;1.73\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4.57\u0026thinsp;\u0026plusmn;\u0026thinsp;1.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;0.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.782\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCa (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.32\u0026thinsp;\u0026plusmn;\u0026thinsp;0.31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.28\u0026thinsp;\u0026plusmn;\u0026thinsp;0.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.47\u0026thinsp;\u0026plusmn;\u0026thinsp;0.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -6.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eP (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.87\u0026thinsp;\u0026plusmn;\u0026thinsp;0.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.82\u0026thinsp;\u0026plusmn;\u0026thinsp;0.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.03\u0026thinsp;\u0026plusmn;\u0026thinsp;0.29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -3.44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eALP (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, U/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e359.47\u0026thinsp;\u0026plusmn;\u0026thinsp;202.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e346.60\u0026thinsp;\u0026plusmn;\u0026thinsp;213.50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e397.89\u0026thinsp;\u0026plusmn;\u0026thinsp;159.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -1.72\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.086\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAST (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, U/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e97.19\u0026thinsp;\u0026plusmn;\u0026thinsp;138.10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e94.18\u0026thinsp;\u0026plusmn;\u0026thinsp;150.51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e106.16\u0026thinsp;\u0026plusmn;\u0026thinsp;92.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -0.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.559\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eALT (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, U/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e49.55\u0026thinsp;\u0026plusmn;\u0026thinsp;73.36\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e47.57\u0026thinsp;\u0026plusmn;\u0026thinsp;79.88\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e55.48\u0026thinsp;\u0026plusmn;\u0026thinsp;49.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -0.73\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.467\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAST/ALT (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2.53\u0026thinsp;\u0026plusmn;\u0026thinsp;2.20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.63\u0026thinsp;\u0026plusmn;\u0026thinsp;2.50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.25\u0026thinsp;\u0026plusmn;\u0026thinsp;0.81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;1.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.254\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGGT (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, U/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e318.49\u0026thinsp;\u0026plusmn;\u0026thinsp;306.19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e210.36\u0026thinsp;\u0026plusmn;\u0026thinsp;183.84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e641.09\u0026thinsp;\u0026plusmn;\u0026thinsp;367.30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -8.80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTBIL (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, umol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e137.54\u0026thinsp;\u0026plusmn;\u0026thinsp;86.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e117.34\u0026thinsp;\u0026plusmn;\u0026thinsp;80.17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e197.79\u0026thinsp;\u0026plusmn;\u0026thinsp;78.41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -6.82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDBIL (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, umol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e62.41\u0026thinsp;\u0026plusmn;\u0026thinsp;55.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e52.80\u0026thinsp;\u0026plusmn;\u0026thinsp;52.23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e91.07\u0026thinsp;\u0026plusmn;\u0026thinsp;54.93\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -4.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIBIL (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, umol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e75.13\u0026thinsp;\u0026plusmn;\u0026thinsp;56.13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e64.54\u0026thinsp;\u0026plusmn;\u0026thinsp;46.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e106.72\u0026thinsp;\u0026plusmn;\u0026thinsp;69.80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -4.41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eD/I (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.35\u0026thinsp;\u0026plusmn;\u0026thinsp;2.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.15\u0026thinsp;\u0026plusmn;\u0026thinsp;1.43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.97\u0026thinsp;\u0026plusmn;\u0026thinsp;5.24\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -1.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.228\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTBA (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, umol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e67.82\u0026thinsp;\u0026plusmn;\u0026thinsp;60.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e51.21\u0026thinsp;\u0026plusmn;\u0026thinsp;44.56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e117.38\u0026thinsp;\u0026plusmn;\u0026thinsp;75.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -6.51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMg (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.88\u0026thinsp;\u0026plusmn;\u0026thinsp;0.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.86\u0026thinsp;\u0026plusmn;\u0026thinsp;0.17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.94\u0026thinsp;\u0026plusmn;\u0026thinsp;0.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -5.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTG (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.22\u0026thinsp;\u0026plusmn;\u0026thinsp;1.47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.29\u0026thinsp;\u0026plusmn;\u0026thinsp;1.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.01\u0026thinsp;\u0026plusmn;\u0026thinsp;0.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;1.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.212\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCK (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, U/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e225.24\u0026thinsp;\u0026plusmn;\u0026thinsp;512.95\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e243.76\u0026thinsp;\u0026plusmn;\u0026thinsp;584.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e169.98\u0026thinsp;\u0026plusmn;\u0026thinsp;165.86\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;0.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.332\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLDH (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, U/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e521.27\u0026thinsp;\u0026plusmn;\u0026thinsp;561.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e572.96\u0026thinsp;\u0026plusmn;\u0026thinsp;634.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e367.05\u0026thinsp;\u0026plusmn;\u0026thinsp;152.58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;4.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHDL-C (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.76\u0026thinsp;\u0026plusmn;\u0026thinsp;0.39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.75\u0026thinsp;\u0026plusmn;\u0026thinsp;0.42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.82\u0026thinsp;\u0026plusmn;\u0026thinsp;0.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -1.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.107\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLDL-C (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.55\u0026thinsp;\u0026plusmn;\u0026thinsp;1.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.41\u0026thinsp;\u0026plusmn;\u0026thinsp;0.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.00\u0026thinsp;\u0026plusmn;\u0026thinsp;1.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et = -3.94\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVLDL-C (Mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD, mmol/L)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.40\u0026thinsp;\u0026plusmn;\u0026thinsp;0.55\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.41\u0026thinsp;\u0026plusmn;\u0026thinsp;0.58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.37\u0026thinsp;\u0026plusmn;\u0026thinsp;0.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003et\u0026thinsp;=\u0026thinsp;0.51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.607\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003eAbbreviations: t, t-test; χ\u0026sup2;, Chi-square test; SD, standard deviation; \u003cem\u003eP\u003c/em\u003e, P-value; K, potassium; Na, sodium; Cl, chloride; CO₂, carbon dioxide; AG, anion gap; OSM, osmolality; TP, total protein; PA, prealbumin; ALB, albumin; GLO, globulin; A/G, albumin/globulin ratio; Urea, urea; Cr, creatinine; Urea/Cr, urea/creatinine ratio; Chol, total cholesterol; UA, uric acid; GLU, glucose; Ca, calcium; P, phosphorus; ALP, alkaline phosphatase; AST, aspartate aminotransferase; ALT, alanine aminotransferase; AST/ALT, AST/ALT ratio; GGT, gamma-glutamyl transferase; TBIL, total bilirubin; DBIL, direct bilirubin; IBIL, indirect bilirubin; D/I, DBIL/ IBIL ratio; TBA, total bile acids; Mg, magnesium; TG, triglycerides; CK, creatine kinase; LDH, lactate dehydrogenase; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; VLDL-C, very-low-density lipoprotein cholesterol.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Univariate and Multivariate Predictors of Neonatal BA\u003c/h2\u003e\u003cp\u003eUnivariable analysis identified 11 BA predictors (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05), among which GGT (OR\u0026thinsp;=\u0026thinsp;1.01, 95%CI: 1.01\u0026ndash;1.01), TBIL (OR\u0026thinsp;=\u0026thinsp;1.01, 95%CI: 1.01\u0026ndash;1.01), and male sex (protective: OR\u0026thinsp;=\u0026thinsp;0.30, 95%CI: 0.16\u0026ndash;0.55) showed strongest associations. Multivariable analysis established five independent predictors: GGT (OR\u0026thinsp;=\u0026thinsp;1.01, 95%CI: 1.01\u0026ndash;1.01, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), TBIL (OR\u0026thinsp;=\u0026thinsp;1.01, 95%CI: 1.01\u0026ndash;1.01, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.010), albumin (ALB: OR\u0026thinsp;=\u0026thinsp;1.17, 95%CI: 1.03\u0026ndash;1.32, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.014), total bile acids (TBA: OR\u0026thinsp;=\u0026thinsp;1.02, 95%CI: 1.01\u0026ndash;1.02, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and urea (Urea: OR\u0026thinsp;=\u0026thinsp;0.75, 95%CI: 0.57\u0026ndash;0.99, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.041; Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003e༎Univariate and Multivariate Analysis of Neonatal BA Diagnostic Characteristics\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"12\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eVariables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"5\" nameend=\"c6\" namest=\"c2\"\u003e\u003cp\u003eUnivariate Analysis\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"5\" nameend=\"c12\" namest=\"c8\"\u003e\u003cp\u003eMultivariate Analysis\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eβ\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eS.E\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eZ\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eOR (95%CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003eβ\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c9\"\u003e\u003cp\u003eS.E\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c10\"\u003e\u003cp\u003eZ\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c11\"\u003e\u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c12\"\u003e\u003cp\u003eOR (95%CI)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-0.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.799\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.00 (0.96\u0026ndash;1.03)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGender\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFemale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.00 (Reference)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMale\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-1.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-3.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.30 (0.16\u0026ndash;0.55)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e3.30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.07 (1.03\u0026ndash;1.11)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.191\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.00 (1.00-1.01)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eALB\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e4.86\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.18 (1.11\u0026ndash;1.27)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e2.47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e\u003cp\u003e0.014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e\u003cp\u003e1.17 (1.03\u0026ndash;1.32)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eUrea\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-3.82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.71 (0.60\u0026ndash;0.85)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e-0.29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e-2.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e\u003cp\u003e0.041\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e\u003cp\u003e0.75 (0.57\u0026ndash;0.99)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTBIL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.01 (1.01\u0026ndash;1.01)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e2.56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e\u003cp\u003e0.010\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e\u003cp\u003e1.01 (1.01\u0026ndash;1.01)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDBIL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e4.17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.01 (1.01\u0026ndash;1.02)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAST\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.560\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.00 (1.00\u0026ndash;1.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGGT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.01 (1.01\u0026ndash;1.01)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e5.08\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e\u003cp\u003e1.01 (1.01\u0026ndash;1.01)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eALP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.68\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.093\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.00 (1.00\u0026ndash;1.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTBA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e5.45\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.02 (1.02\u0026ndash;1.03)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e3.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c12\"\u003e\u003cp\u003e1.02 (1.01\u0026ndash;1.02)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTG\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-1.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.207\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.77 (0.52\u0026ndash;1.15)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLDH\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-2.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.99 (0.99\u0026ndash;0.99)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHDL-C\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.49\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.201\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.63 (0.77\u0026ndash;3.45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLDL-C\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.55\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e3.60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.73 (1.28\u0026ndash;2.34)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVLDL-C\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e-0.51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.607\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.86 (0.48\u0026ndash;1.54)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"12\"\u003eAbbreviations: β, beta coefficient; S.E, standard error; Z, Z-value; P, P-value; OR (95%CI), odds ratio (95% confidence interval); TP, total protein; PA, prealbumin; ALB, albumin; Urea, urea; TBIL, total bilirubin; DBIL, direct bilirubin; AST, aspartate aminotransferase; GGT, gamma-glutamyl transferase; ALP, alkaline phosphatase; TBA, total bile acids; TG, triglycerides; LDH, lactate dehydrogenase; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; VLDL-C, very-low-density lipoprotein cholesterol.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Machine Learning-Based Diagnostic Screening Models for Neonatal BA\u003c/h2\u003e\u003cp\u003eSerum biomarkers from 243 neonates were used to construct five machine learning-based diagnostic screening models for early BA detection. Feature selection was validated by Spearman correlation (r\u0026thinsp;\u0026lt;\u0026thinsp;0.9), with LASSO regression identifying six optimal predictors: ALB, GGT, TBIL, TBA, gender, and urea (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and Supplementary Fig.\u0026nbsp;1). Following 10-fold cross-validation and Optuna hyperparameter tuning, all models demonstrated robust training performance (AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.8; Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). During training, support vector classifier (SVC; AUC\u0026thinsp;=\u0026thinsp;0.962), and random forest (RF; AUC\u0026thinsp;=\u0026thinsp;0.952) achieved the highest discriminative power. In the independent validation set, RF attained best classification (AUC\u0026thinsp;=\u0026thinsp;0.915, 95%CI: 0.827-1.000; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). RF exhibited exceptional reproducibility, with consistent AUCs in training (0.959) and validation (0.915) phases. Calibration curves confirmed high agreement between predicted and observed probabilities across all models (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDiagnostic screening performance of five machine learning models for neonatal biliary atresia in the training and validation sets\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"12\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eAlgorithms\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c6\" namest=\"c2\"\u003e\u003cp\u003eAverage performance of cross-validation on the training set\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\" morerows=\"6\" rowspan=\"7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c12\" namest=\"c8\"\u003e\u003cp\u003ePerformance on the validation set\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePrecision\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eRecall\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eF1-score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eAccuracy\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eAUROC (95%CI)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003ePrecision\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003eRecall\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003eF1-score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003eAccuracy\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003eAUROC (95%CI)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLRM\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.867\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.455\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.578\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.853\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.937 (0.900-0.973)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.500\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.667\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0.863\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e0.750 (0.614\u0026ndash;0.886)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDET\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.708\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.685\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.678\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.853\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.796 (0.721\u0026ndash;0.870)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.621\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.900\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.735\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0.822\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e0.846 (0.733\u0026ndash;0.960)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMLP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.750\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.425\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.504\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.829\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.868 (0.815\u0026ndash;0.921)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.667\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.400\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.500\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0.781\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e0.662 (0.516\u0026ndash;0.809)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSVC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.855\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.705\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.743\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.894\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.962 (0.940\u0026ndash;0.985)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.750\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.750\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.750\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0.890\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e0.828 (0.709\u0026ndash;0.947)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRF\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.843\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.805\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.815\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.912\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.952 (0.917\u0026ndash;0.987)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.690\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.816\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c11\"\u003e\u003cp\u003e0.877\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c12\"\u003e\u003cp\u003e0.915 (0.827-1.000)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"12\"\u003eAbbreviation: AUROC, area under the receiver operating characteristic; CI, confidence interval; ML, machine learning; LRM, logistic regression model; RF, random forest; SVC, support vector machine classifier; MLP, multilayer perceptron; DET, decision tree\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e3.4 Clinical Deployment of the Neonatal BA Diagnostic Screening Platform\u003c/h2\u003e\u003cp\u003eConsistent feature importance analysis using SHAP summary plots (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA-F) established GGT, TBIL, and TBA as the most influential predictors across all five diagnostic screening models. SHAP dependence plots further elucidated biomarker thresholds and feature interactions, revealing monotonic relationships where elevated values (represented by red data points in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA-F) consistently increased BA probability. These computational insights drove the selection of the random forest model - distinguished by best validation performance (AUC\u0026thinsp;=\u0026thinsp;0.915, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e) - for clinical implementation as a web-based diagnostic screening tool.\u003c/p\u003e\u003cp\u003eThe deployed interface (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA-B) features an intuitive input panel for six parameters: ALB (g/L), GGT (U/L), TBIL (\u0026micro;mol/L), TBA (\u0026micro;mol/L), sex (categorical), and urea (mmol/L). Upon submission, the platform generates comprehensive clinical outputs in \u0026lt;\u0026thinsp;2 seconds: (i) binary BA/non-BA classification, (ii) quantified risk probability (0-100%) stratified as low (\u0026lt;\u0026thinsp;20%), intermediate (20\u0026ndash;80%), or high (\u0026gt;\u0026thinsp;80%), and (iii) an interpretable visualization of feature contributions modeled through SHAP force plots. Implementation resources including cross-platform installation packages are detailed in Supplementary Files.\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eIn this study, we developed and validated a ML-based screening model using routine biochemical parameters and basic demographic data to assist in the early identification of BA among cholestatic neonates. The model exhibited excellent discriminative performance, with several algorithms, particularly RF and SVC, achieving near-perfect classification in the validation cohort. Furthermore, the model was successfully deployed as a web-based interface, enabling real-time clinical application at the bedside.\u003c/p\u003e\u003cp\u003eGGT was identified as the most influential predictor across all machine learning models. Elevated GGT levels are known to be strongly associated with BA and reflect cholangiocyte proliferation and bile duct injury \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. Prior studies have established GGT as a core diagnostic marker for BA, especially in distinguishing it from intrahepatic cholestasis \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. However, previous research has also demonstrated a significant correlation between GGT levels and patient age in BA, suggesting that diagnostic models derived from older infants may not be directly applicable to neonates \u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Our findings reaffirm the pivotal role of GGT in BA diagnosis and support its integration into ML-assisted diagnostic tools.\u003c/p\u003e\u003cp\u003eTBIL and TBA also ranked among the top predictors in the machine learning models. Both are classical biochemical markers of cholestasis, but previous studies have shown that their levels tend to be more markedly elevated in BA compared to other causes of neonatal cholestasis. \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. In our SHAP analysis, higher TBIL and TBA levels were associated with an increased predicted probability of BA, demonstrating a clear monotonic relationship. Although these indicators are routinely measured in clinical settings, their integration into ML models significantly improved diagnostic performance. These findings highlight the potential of combining conventional biochemical markers with advanced algorithms to enhance diagnostic accuracy in neonatal BA.\u003c/p\u003e\u003cp\u003eInterestingly, ALB levels were significantly higher in the BA group compared to the non-BA group in our study. While hypoalbuminemia is typically associated with impaired hepatic function, this finding suggests that neonates with BA may be in an early disease stage, during which albumin synthesis remains preserved or even upregulated. This paradoxical elevation may reflect compensatory hepatic synthetic activity or early nutritional support. In contrast, non-BA cholestasis, such as parenteral nutrition-associated cholestasis, was often accompanied by acute hepatocellular injury, which can lead to early reductions in albumin production\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Incorporating ALB into our model enhanced its ability to capture early functional differences in the liver, supporting its utility as an auxiliary marker in neonatal BA screening. Serum urea levels were negatively associated with BA. Lower urea levels in BA infants may reflect reduced protein catabolism or early shifts in hepatic nitrogen metabolism.\u003c/p\u003e\u003cp\u003eFemale sex was also identified as an independent risk factor for BA, consistent with previous studies reporting a higher incidence of BA among female infants\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Choi et al. incorporated demographic variables such as age and sex, along with biochemical markers, into their diagnostic model for BA, achieving an AUC of 0.97 \u003csup\u003e26\u003c/sup\u003e. Similarly, our study found that including sex as a variable improved the overall predictive performance of the model, highlighting the value of integrating demographic data into early diagnostic strategies.\u003c/p\u003e\u003cp\u003ePrevious ML-based studies on BA have frequently incorporated advanced imaging features, liver stiffness measurements, or novel biomarkers such as MMP-7 \u003csup\u003e5, 15, 27\u003c/sup\u003e. Among these, MMP-7 has emerged as a promising serological marker with high diagnostic accuracy, showing AUC values consistently above 0.90 in multiple studies \u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. However, several limitations hinder its integration into routine clinical practice, including high testing cost, limited accessibility in many hospitals, and lack of standardized cutoff values across age groups. In contrast, our ML model is built entirely upon routine biochemical parameters that are universally available, inexpensive, and already part of standard neonatal cholestasis workups. Therefore, the simplicity, reproducibility, and affordability of our model position it as a practical alternative for early BA screening, especially in resource-constrained regions.\u003c/p\u003e\u003cp\u003eThe primary strengths of our study include its data-driven feature selection, rigorous model validation, and clinically deployable web interface. However, this work has some limitations. First, it was based on a single-center retrospective cohort, which may limit generalizability. External validation using multicenter datasets is needed. Second, the model excluded imaging features that may offer additional diagnostic value. Third, real-world clinical impact of the online tool requires prospective testing in diverse practice settings.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eIn conclusion, this study developed diagnostic screening models for neonatal biliary atresia using routine biochemical markers and demographic characteristics. The random forest model showed the highest predictive accuracy and may serve as a practical screening tool to aid in the early identification of BA. Although further external validation is required, these findings suggest that machine learning-based diagnostic screening models hold promise for improving the diagnostic pathway and clinical management of neonatal BA.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003e6.1 Acknowledgments\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe authors would like to thank all the medical and nursing staff from the Department of Neonatal Surgery, Beijing Children\u0026rsquo;s Hospital, Capital Medical University, for their dedicated patient care and assistance in data collection. We are also grateful to the Central Laboratory of Jiangxi Provincial Children\u0026rsquo;s Hospital for providing technical support during this study.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e6.2 Funding\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThis work was supported by the Beijing Municipal Natural Science Foundation (Grant No. 7252043 to JSH), the Jiangxi (Ganpo) Talent Program-Health Innovation Talent Project (to JSH), and the National Natural Science Foundation of China (Grant No. 82400592 to JDW; Grant No. 82300574 to DYS).\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e6.3 Conflict of Interest\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe authors declare no conflicts of interest related to this publication.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e6.4 Author Contributions\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy conception and design (JSH, DDW); data acquisition (ZZL, YYJ, YZ, YNZ, SSL, JML, KYH, YCG, DYS); data analysis and interpretation (ZZL, YYJ, JSH, DDW); drafting of the manuscript (ZZL, YYJ); critical revision of the manuscript for important intellectual content (JSH, JDW). All authors have read and approved the final manuscript. Zhaozhou Liu and Yuyan Jin contributed equally as co-first authors.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e6.5 Ethical Statement\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThis study was approved by the Ethics Committee of Beijing Children\u0026rsquo;s Hospital, Capital Medical University (Approval No. 2019-k-386). The requirement for informed consent was waived due to the retrospective and anonymized design of the study. The research was conducted in accordance with the principles of the Declaration of Helsinki.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003e6.6 Data Sharing Statement\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request. No additional data are available.\u003c/strong\u003e\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eChung PHY, Zheng S, Tam PKH. Biliary atresia: East versus west. \u003cem\u003eSeminars in pediatric surgery.\u003c/em\u003e 2020;29:150950.\u003c/li\u003e\n\u003cli\u003eTam PKH, Wells RG, Tang CSM, et al. Biliary atresia. \u003cem\u003eNature reviews. Disease primers.\u003c/em\u003e 2024;10:47.\u003c/li\u003e\n\u003cli\u003eFanna M, Masson G, Capito C, et al. Management of Biliary Atresia in France 1986 to 2015: Long-term Results. \u003cem\u003eJournal of pediatric gastroenterology and nutrition.\u003c/em\u003e 2019;69:416-424.\u003c/li\u003e\n\u003cli\u003eXie C, Wang P, Wang D, et al. Is performing the Kasai portoenterostomy in the neonatal period associated with a better prognosis? A single-center, retrospective cohort study from China. \u003cem\u003eBMC pediatrics.\u003c/em\u003e 2025;25:454.\u003c/li\u003e\n\u003cli\u003eHou J, Xiao W, Zhou S, Liu H. Identification of Biliary Atresia in Infantile Cholestasis: Integrating Radiomics With MRCP for Unobservable Extrahepatic Biliary Systems. \u003cem\u003eJournal of computer assisted tomography.\u003c/em\u003e 2025.\u003c/li\u003e\n\u003cli\u003eKianifar HR, Tehranian S, Shojaei P, et al. Accuracy of hepatobiliary scintigraphy for differentiation of neonatal hepatitis from biliary atresia: systematic review and meta-analysis of the literature. \u003cem\u003ePediatric radiology.\u003c/em\u003e 2013;43:905-919.\u003c/li\u003e\n\u003cli\u003eMa Y, Yang Y, Du Y, et al. Development of an artificial intelligence-based multimodal diagnostic system for early detection of biliary atresia. \u003cem\u003eBMC medicine.\u003c/em\u003e 2025;23:127.\u003c/li\u003e\n\u003cli\u003eBrahee DD, Lampl BS. Neonatal diagnosis of biliary atresia: a practical review and update. \u003cem\u003ePediatric radiology.\u003c/em\u003e 2022;52:685-692.\u003c/li\u003e\n\u003cli\u003eHarpavat S, Garcia-Prats JA, Anaya C, et al. Diagnostic Yield of Newborn Screening for Biliary Atresia Using Direct or Conjugated Bilirubin Measurements. \u003cem\u003eJAMA.\u003c/em\u003e 2020;323:1141-1150.\u003c/li\u003e\n\u003cli\u003eGong Z, Lin L, Lu G, Wan C. Development and validation of a model for early diagnosis of biliary atresia. \u003cem\u003eBMC pediatrics.\u003c/em\u003e 2023;23:549.\u003c/li\u003e\n\u003cli\u003eGolden J, Zagory JA, Fenlon M, et al. Liquid chromatography-mass spectroscopy in the diagnosis of biliary atresia in children with hyperbilirubinemia. \u003cem\u003eThe Journal of surgical research.\u003c/em\u003e 2018;228:228-237.\u003c/li\u003e\n\u003cli\u003ePandurangi S, Mourya R, Nalluri S, et al. Diagnostic accuracy of serum matrix metalloproteinase-7 as a biomarker of biliary atresia in a large North American cohort. \u003cem\u003eHepatology (Baltimore, Md.).\u003c/em\u003e 2024;80:152-162.\u003c/li\u003e\n\u003cli\u003eFu M, Guo Z, Chen Y, et al. Proteomics Defines Plasma Biomarkers for the Early Diagnosis of Biliary Atresia. \u003cem\u003eJournal of proteome research.\u003c/em\u003e 2024;23:1744-1756.\u003c/li\u003e\n\u003cli\u003eLyu H, Ye Y, Lui VCH, et al. Plasma amyloid-beta levels correlated with impaired hepatic functions: An adjuvant biomarker for the diagnosis of biliary atresia. \u003cem\u003eFrontiers in surgery.\u003c/em\u003e 2022;9:931637.\u003c/li\u003e\n\u003cli\u003eZhou W, Yang Y, Yu C, et al. Ensembled deep learning model outperforms human experts in diagnosing biliary atresia from sonographic gallbladder images. \u003cem\u003eNature communications.\u003c/em\u003e 2021;12:1259.\u003c/li\u003e\n\u003cli\u003eHsu F-R, Dai S-T, Chou C-M, Huang S-Y. The application of artificial intelligence to support biliary atresia screening by ultrasound images: A study based on deep learning models. \u003cem\u003ePloS one.\u003c/em\u003e 2022;17:e0276278.\u003c/li\u003e\n\u003cli\u003eSun Y, Dai S, Shen Z, et al. Gamma-glutamyl transpeptidase has different efficacy on biliary atresia diagnosis in different hospital patient groups: an application of machine learning approach. \u003cem\u003ePediatric surgery international.\u003c/em\u003e 2022;38:1131-1141.\u003c/li\u003e\n\u003cli\u003eZhao Y, Wang A, Wang D, et al. Development of a diagnostic model for biliary atresia based on MMP7 and serological tests using machine learning. \u003cem\u003ePediatric surgery international.\u003c/em\u003e 2024;40:203.\u003c/li\u003e\n\u003cli\u003eKong F, Dong R, Chen G, et al. Progress in Biomarkers Related to Biliary Atresia. \u003cem\u003eJournal of clinical and translational hepatology.\u003c/em\u003e 2024;12:305-315.\u003c/li\u003e\n\u003cli\u003eHsiao Y-H, Hung W-L, Yang Y-J, et al. Non-invasive simple predictors of biliary atresia in cholestatic infants - A preliminary report. \u003cem\u003ePediatrics and neonatology.\u003c/em\u003e 2025.\u003c/li\u003e\n\u003cli\u003eChen X, Dong R, Shen Z, Yan W, Zheng S. Value of Gamma-Glutamyl Transpeptidase for Diagnosis of Biliary Atresia by Correlation With Age. \u003cem\u003eJournal of pediatric gastroenterology and nutrition.\u003c/em\u003e 2016;63:370-373.\u003c/li\u003e\n\u003cli\u003eHan Y-J, Hu S-Q, Zhu J-H, et al. Accurate prediction of biliary atresia with an integrated model using MMP-7 levels and bile acids. \u003cem\u003eWorld journal of pediatrics : WJP.\u003c/em\u003e 2024;20:822-833.\u003c/li\u003e\n\u003cli\u003eFeldman AG, Sokol RJ. Neonatal Cholestasis: Updates on Diagnostics, Therapeutics, and Prevention. \u003cem\u003eNeoReviews.\u003c/em\u003e 2021;22:e819-e836.\u003c/li\u003e\n\u003cli\u003eLee KJ, Kim JW, Moon JS, Ko JS. Epidemiology of Biliary Atresia in Korea. \u003cem\u003eJournal of Korean medical science.\u003c/em\u003e 2017;32:656-660.\u003c/li\u003e\n\u003cli\u003eHukkinen M, Kerola A, Lohi J, et al. Treatment Policy and Liver Histopathology Predict Biliary Atresia Outcomes: Results after National Centralization and Protocol Biopsies. \u003cem\u003eJournal of the American College of Surgeons.\u003c/em\u003e 2018;226:46-57.e41.\u003c/li\u003e\n\u003cli\u003eChoi HJ, Kim YE, Namgoong J-M, et al. Development and Validation of a Machine Learning-Based Prediction Model for Detection of Biliary Atresia. \u003cem\u003eGastro hep advances.\u003c/em\u003e 2023;2:778-787.\u003c/li\u003e\n\u003cli\u003eChen X, Zhao D, Ji H, Chen Y, Li Y, Zuo Z. Predictive modeling for early detection of biliary atresia in infants with cholestasis: Insights from a machine learning study. \u003cem\u003eComputers in biology and medicine.\u003c/em\u003e 2024;174:108439.\u003c/li\u003e\n\u003cli\u003eZhang Y, Li T, Wang T, Ji Q, Zhan J. Comparison for the diagnostic performance of early diagnostic methods for biliary atresia: a systematic review and network meta-analysis. \u003cem\u003ePediatric surgery international.\u003c/em\u003e 2024;40:146.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"pediatric-surgery-international","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pesi","sideBox":"Learn more about [Pediatric Surgery International](http://link.springer.com/journal/383)","snPcode":"383","submissionUrl":"https://submission.nature.com/new-submission/383/3","title":"Pediatric Surgery International","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Biliary atresia, Neonate, Diagnosis, Machine learning","lastPublishedDoi":"10.21203/rs.3.rs-7460903/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7460903/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eBiliary atresia (BA) is a progressive neonatal cholestatic liver disease that requires timely diagnosis and intervention. Differentiating BA from other causes of neonatal cholestasis remains a significant clinical challenge.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eIn this study, we retrospectively analyzed the clinical and biochemical data of 243 cholestatic neonates, comprising 61 with BA and 182 with non-BA. We utilized five supervised machine learning algorithms\u0026mdash;logistic regression (LRM), decision tree (DET), multilayer perceptron (MLP), support vector machine (SVC), and random forest (RF)\u0026mdash;to construct diagnostic models for BA. The performance of each model was evaluated based on its accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). We then developed an online diagnostic tool based on the best-performing model.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eThe BA and non-BA groups showed significant differences across multiple biochemical markers. All five models demonstrated good diagnostic performance, with the random forest (RF) model achieving the best results (AUC\u0026thinsp;=\u0026thinsp;0.93, sensitivity\u0026thinsp;=\u0026thinsp;88.5%, specificity\u0026thinsp;=\u0026thinsp;85.2%). The combination of multiple biochemical parameters substantially improved diagnostic accuracy compared to using single indicators. The web-based tool provides an intuitive and user-friendly interface to support early BA screening in clinical practice.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eMachine learning-based models, particularly the RF model, show great potential for the early diagnosis of BA in cholestatic neonates. The implementation of a dedicated online platform may facilitate timely identification and assist clinicians in decision-making.\u003c/p\u003e","manuscriptTitle":"Development and Validation of a Screening Model for Early Diagnosis of Biliary Atresia in Neonates with Cholestasis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-09 11:58:32","doi":"10.21203/rs.3.rs-7460903/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-16T15:54:15+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-13T20:33:08+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"160250180173592595464867601185239817076","date":"2025-09-22T06:43:55+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-02T07:36:21+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-29T06:12:42+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-28T07:44:10+00:00","index":"","fulltext":""},{"type":"submitted","content":"Pediatric Surgery International","date":"2025-08-26T08:43:29+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"pediatric-surgery-international","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pesi","sideBox":"Learn more about [Pediatric Surgery International](http://link.springer.com/journal/383)","snPcode":"383","submissionUrl":"https://submission.nature.com/new-submission/383/3","title":"Pediatric Surgery International","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"717777ee-2462-4af5-861c-6b9df88a08f5","owner":[],"postedDate":"September 9th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-02-10T09:56:26+00:00","versionOfRecord":{"articleIdentity":"rs-7460903","link":"https://doi.org/10.1007/s00383-025-06258-6","journal":{"identity":"pediatric-surgery-international","isVorOnly":false,"title":"Pediatric Surgery International"},"publishedOn":"2025-12-20 15:57:44","publishedOnDateReadable":"December 20th, 2025"},"versionCreatedAt":"2025-09-09 11:58:32","video":"","vorDoi":"10.1007/s00383-025-06258-6","vorDoiUrl":"https://doi.org/10.1007/s00383-025-06258-6","workflowStages":[]},"version":"v1","identity":"rs-7460903","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7460903","identity":"rs-7460903","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}