Machine learning-based model used for predicting Portal vein thrombosis risk for patients with cirrhosis

Machine learning-based model used for predicting Portal vein thrombosis risk for patients with cirrhosis | 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 Machine learning-based model used for predicting Portal vein thrombosis risk for patients with cirrhosis Peipei Meng, Yang Zhou, Xiaoli Liu, Tong Wu, Hao Yu, Xiaomin Ji, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3831874/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract We aimed to assess the risk of portal vein thrombosis (PVT) in patients with hepatitis B-related cirrhosis (HBC) using artificial neural networks (ANN). PVT can exacerbate portal hypertension and lead to complications, increasing the risk of mortality. Unfortunately, accurate prediction models for PVT in hepatitis B cirrhosis patients are currently insufficient. To address this gap, we conducted a study at Beijing Ditan Hospital, affiliated with Capital Medical University, involving 986 hospitalized patients. The patients were randomly divided into a training set (685 cases) and a validation set (301 cases) using a 3:1 ratio. Through univariate analysis, we determined independent factors that influence the occurrence of PVT, which were then utilized to develop an ANN model. The performance of the ANN model was assessed using various indicators, such as the area under the receiver operating characteristic curve (AUC) and concordance index (C-index). In the training group, PVT developed within three years in 19.0% of patients, and within five years in 23.7% of patients. Similarly, in the validation group, PVT developed within three years in 16.7% of patients, and within five years in 24.0% of patients. The ANN model incorporated nine independent risk factors, including age, presence of ascites, manifestation of hepatic encephalopathy (HE), occurrence of gastrointestinal varices with bleeding, Child-Pugh classification, alanine transaminase (ALT) levels, albumin (ALB) levels, neutrophil-to-lymphocyte ratio (NLR), and platelet count (PLT). Importantly, the AUC of the ANN model was significantly higher at 0.9718 compared to existing models such as MELD and CTP (all p<0.001). Our ANN model effectively classified patients into high ,medium, and low risk groups for PVT development over a span of 3 and 5 years. These findings were further validated in an independent cohort. Machine learning-based model Portal vein thrombosis Risk Hepatitis B-related cirrhosis Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Portal Vein Thrombosis (PVT) is a type of thrombosis that occurs in the main portal vein and its branches, which can result in partial or complete blockage of the blood vessel. Traditionally, it has been believed that individuals with cirrhosis of the liver experience clotting factor synthesis disorders and thrombocytopenia, making them more prone to bleeding. However, recent research suggests that the anticoagulant and procoagulant systems in the liver of individuals with cirrhosis are in a dynamic and unstable state, leading to stagnation of blood flow in the portal vein and dysfunction of the blood vessel lining. This can increase the risk of both bleeding and thrombosis ( 1 , 2 ). In fact, non-tumoral PVT is relatively common in patients with cirrhosis, with an estimated annual incidence ranging from 4.6–26% ( 2 , 3 ), and higher in those with more advanced liver disease. Currently, most studies on PVT analyze influencing factors using retrospective data, and there is no reliable model for early prediction of PVT occurrence. As a result, identifying high-risk individuals for early prevention is challenging in clinical practice ( 4 , 5 , 6 , 7 , 8 ). Some studies have suggested that a portal blood flow velocity (PBFV) below 15 cm/sec ( 4 , 9 ) is a major risk factor for PVT development. Other risk factors include the severity of liver disease and the presence of portal hypertension, such as low platelet count ( 4 , 6 , 10 ), low albumin levels ( 11 ), large esophageal varices ( 3 , 7 , 12 ), previous sclerotherapy ( 13 ), previous liver decompensation ( 14 ), and the presence of large portosystemic collaterals ( 15 ). More recently, it has been proposed that the Model for End-Stage Liver Disease (MELD) and Child-Pugh-Turcotte (CTP) scores can help predict the development of PVT ( 7 , 12 , 16 ). However, these studies did not adequately consider the potential influence of other recognized risk factors for PVT Artificial Neural Networks (ANNs) are a form of machine learning that mimics the information processing of brain neurons ( 17 , 18 , 19 , 20 ). These mathematical models have found extensive applications in medical decision-making by analyzing linear, logistic, and nonlinear complex relationships ( 21 ). Through training, ANNs optimize the factors associated with the outcome, resulting in highly accurate prediction models. Therefore, this study aims to employ ANNs in constructing an advanced warning model for accurate identification of high-risk groups prone to disease progression, specifically predicting the PVT Materials and methods 2.1 Patients A total of 1505 patients diagnosed with cirrhosis were retrospectively enrolled at the Beijing Ditan Hospital of Capital Medical University in Beijing, China, from January 2011 to January 2016. Eligibility criteria included being a first-time diagnosis of cirrhosis, age between 18 and 80 years, and confirmation of cirrhosis through liver biopsy and/or compatible clinical, laboratory, and imaging data. Patients who died within a 3-year or 5-year period or were lost to follow-up were excluded. Exclusion criteria encompassed individuals with known hepatocellular carcinoma (HCC), pregnant women, those who have undergone previous orthotopic liver transplantation (OLT), unwillingness to provide informed consent, and use of anticoagulation therapy or prior surgical or transjugular intrahepatic portosystemic shunt (TIPS) procedures. To ensure a representative sample, 986 patients were randomly assigned, with 70% (n = 685) allocated to the training cohort and the remaining 30% (n = 301) assigned to the validation cohort (Fig. 1 ). This study received ethical approval from the Ethics Committee of Beijing Ditan Hospita 2.2 Clinical definition and follow-up Compensatory cirrhosis was determined through the following methods: ( 1 ) On biopsy, the presence of pathological findings indicating F4 stage cirrhosis; ( 2 ) During endoscopy, the presence of esophageal varices and exclusion of noncirrhotic portal hypertension; ( 3 ) In the absence of histological evidence and endoscopic findings, at least two out of three criteria should be met:①Imaging techniques such as ultrasonography, computed tomography, or magnetic resonance imaging show changes in liver morphology, such as nodules in the liver tissue and uneven texture on the liver surface; ②Platelet count lower than 100×10 9 cells/L, without any other identifiable causes; ③Serum albumin lower than 35.0g/L, international normalized ratio higher than 1.3, or prothrombin time prolonged by more than 3 seconds. The diagnosis of decompensated cirrhosis is based on the presence of cirrhosis along with complications related to portal and venous hypertension and/or liver dysfunction. ( 1 ) The diagnosis requires evidence of cirrhosis; ( 2 ) Presence of complications associated with portal hypertension such as ascites, bleeding from esophageal and gastric varices, and hepatic encephalopathy ( 22 ). PVT (portal vein thrombosis) is diagnosed by the visualization of non-tumoral thrombosis within the portal vein or its branches. Such diagnosis and assessment of its extent are always confirmed through computed tomography or magnetic resonance imaging. PVT is categorized as occlusive if there is a complete absence of blood flow in the vein, and partial if the lumen is only partially occluded while blood flow is still present The starting point of this research was the initial cirrhosis diagnosis at the hospital. The conclusion of this study focused on the recent identification of PVT within a one-year span and the subsequent follow-up of five years. The collection of clinical data encompassed various categories such as demographics (age, gender), complications (bleeding gastrointestinal varices, ascites, hepatic encephalopathy [HE]), biochemical indicators (alanine aminotransferase [ALT], aspartate aminotransferase [AST], total bilirubin [TBIL], serum albumin [ALB], ɣ-glutamyl transpeptidase [GGT], white blood cell count [WBC], neutrophil count [NC], lymphocyte count [LC], platelet count [PLT], creatinine [CREA], prothrombin time [PT], international normalized ratio [INR], Alpha-fetoprotein [AFP]), and routine laboratory tests which involved computed tomography or magnetic resonance imaging, conducted every 3–6 months 2.3 Construction of ANN The artificial neural network (ANN) is composed of complex and interconnected processing units called neurons. These neurons are connected through weighted connections and organized into an input layer, output layer, and one or more hidden layers ( 23 , 24 ). Some advantages of using an ANN include its ability to self-learn, self-adapt, and perform inference processes. The ANN learns by examining examples and adjusting the weights of the connections between neurons to establish a relationship between input and output. When applied to data, the input is passed through the layers of neurons until an output is generated. Following this output, a process of self-adaptation occurs. The produced output is compared to the desired output, and if there is a difference, an error signal is generated. This error signal is then used in a back propagation (BP) method to modify the weights of the connections between neurons. This modification aims to minimize the overall error of the network. Throughout the learning process, the error between the produced outputs and the desired outputs gradually decreases until it reaches a minimum, indicating convergence of the network. Once this convergence is achieved, the ANN can perform an inference process. During this process, new input data can be used to generate outputs or predictions based on the knowledge gained during the training process. This allows the ANN to accurately predict outcomes on different data sets ( 25 , 26 , 27 ) In this investigation, the 5-year progression of PVT in the 637 layers of cirrhosis input contained neurons that imported the available data, encompassing a variety of clinical, demographic, and laboratory information. The output layers consisted of neurons that exported the corresponding predictive results. The hidden layers were utilized to facilitate intricate interactions between the input and output neurons. Factors significantly linked to PVT in patients with cirrhosis were used to construct artificial neural networks (ANNs) employing Mathematica 11.1.1 for Microsoft Windows (64-bit), a graphical tool for neural network development. A total of 912 patients were assigned to either a training group (n = 637, 70%) or a validation group (n = 275, 30%). The backpropagation (BP) algorithm was employed in the learning process of this ANN, which involved assessing the errors between the generated and desired outputs. The connections between neurons were adjusted by modifying the weights to minimize the overall network errors. The learning (training) process would be terminated once the sum of squared errors reached a minimum compared to the cross-validation dataset. Ultimately, the final model provided the development of PVT risks over a 3/5-year period for each patient. 2.4Statistical analysis The data were expressed in the form of median (range) or n (%) as applicable. To assess the statistical significance of differences among continuous and categorical variables, we employed Student's t-test (or Mann-Whitney test if appropriate) and chi-squared test (or Fisher's exact test if appropriate). Once the relationship between demographic, biochemical, and clinical variables (inputs) and prognosis (outputs) was determined, we selected the variables with statistically significant differences or important clinical characteristics as the input layers for constructing artificial neural networks (ANNs) to predict the development of portal vein thrombosis (PVT) in patients over a period of 3 years and 5 years. We presented hazard ratios (HR) and their corresponding 95% confidence intervals (CI), along with p values. To evaluate the discriminatory performance, we utilized receiver-operating characteristic (ROC) curves. The area under the ROC curve was computed to generate Harrell's c-index. Furthermore, we compared the performance of the ANN model with that of Model for End-Stage Liver Disease (MELD) in the ROC curves [28,29,30]. The scores for MELD were calculated based on the published scoring formula. To visually assess the agreement between the predicted probability of PVT over 3/5 years by the model and the observed probability, we employed a calibration plot. Additionally, we conducted decision curve analysis (DCA) to compare the clinical net benefits of the new model compared to previous models. For all statistical tests, a p-value of less than 0.05 was considered to indicate a statistically significant difference. We performed the statistical analysis using SPSS 22 (IBM, Armonk, NY, USA) and R version 3.3.2 (R core development team, 2010). Results 3.1 Baseline characteristics Between 2008 and 2016, we conducted a study involving the participation of 986 patients. These patients were randomly divided into two datasets, namely the training dataset (n = 685) and the validation dataset (n = 301). Within the three-year period, 121 patients (19.0%) developed PTV in the training group, while 151 patients (23.7%) developed PTV within five years. In the validation group, there were 46 patients (16.7%) who developed PTV within three years and 66 patients (24.0%) who developed PTV within five years. The characteristics of the two groups were comparable as shown in Table 1 . In the training set, a great majority of the participants, specifically 439 individuals (69.0%), were male, and their median age was 52.8 years (interquartile range: 40–74 years). Importantly, no significant differences in baseline characteristics were observed between the two sets(Table 1 ) Table 1 Basic clinical characteristics of patients with HBV-related cirrhosis. Variables All patients (n = 912) Training Cohort (n = 637) Validation Cohort (n = 275) P value Age, years 53.0 (41.0–73.0) 52.8.0 (40.0–74.0) 52.9(43.0–75.0) 0.436 Male sex, n (%) 611 (67.0) 439(69.0) 172(62.5) 0.320 Smoking, n (%) 210(23.0) 149(23.4) 61 (22.2) 0.846 Alcohol consumption, n (%) 195(21.4) 123 (19.3) 62(22.5) 0.579 Diabetes, n (%) 152 (16.7) 103(16.2) 49(17.8) 0.174 Hypertension, n (%) 129 (14.1) 95 (14.9) 34 (12.4) 0.835 Ascites, n (%) 197(21.6) 134 (21.0) 63 (22.9) 0.485 Encephalopathy, n (%) Gastrointestinal varices with bleeding, n (%) 41(4.5) 134(14.7) 30 (4.7) 88(13.8) 11 (4.0) 46(16.7) 0.458 0.348 CTP score 7.0 (5.0–10.0) 7.0 (6.0–9.0) 7.0 (5.0–10.0) 0.626 MELD score 10.1 (7.8–12.9) 10.1 (8.0-12.6) 10.1 (7.7–13.6) 0.851 Alanine aminotransferase, U/L 41.3 (26.5-105.8) 43.6 (27.1-112.4) 37.5 (25.1–95.5) 0.164 Aspartate aminotransferase, U/L 47.9 (30.8-106.1) 48.6 (32.1-107.9) 46.7 (29.7–98.1) 0.282 Total bilirubin, µmol/L 22.4 (14.2–39.2) 22.7 (14.1–38.5) 22.3 (14.5–40.7) 0.805 Albumin, g/L 33.6 (30.3–39.2) 33.8 (30.4–40.1) 33.2 (30.1–38.4) 0.258 Gamma-glutamyl transpeptidase, U/L 56.3 (37.2–97.8) 55.3 (36.1–99.7) 57.3 (40.4–95.6) 0.497 White blood cell count, ×10 9 /L 3.9 (2.8–5.3) 3.9 (2.8–5.3) 3.8 (2.7–5.3) 0.451 Neutrophil count, ×10 9 /L 2.2 (1.5–3.1) 2.2 (1.5–3.2) 2.2 (1.5-3.0) 0.434 Lymphocyte count, ×10 9 /L 1.1 (0.8–1.6) 1.1 (0.8–1.6) 1.1 (0.7–1.6) 0.921 Neutrophil-lymphocyte ratio 2.0 (1.4–2.7) 2.0 (1.4–2.8) 1.9 (1.4–2.6) 0.598 Platelets, ×10 9 /L 87.0 (65.8-118.6) 87.0 (65.0-116.0) 89.0 (67.0-121.0) 0.199 Creatinine, µmol/L 66.0 (56.0–76.0) 66.1 (56.1–76.2) 65.0 (56.0-73.9) 0.602 Blood urea nitrogen, mmol/L 5.1(4.0-6.7) 5.1(4.0-6.7) 5.2 (4.1–6.7) 0.717 Prothrombin time, s 14.3 (12.7–16.3) 14.3 (12.8–16.1) 14.1 (12.5–16.7) 0.801 Prothrombin activity, % 67.0(54.0–80.0) 67.0(54.0–80.0) 67.0(53.0–81.0) 0.944 International normalized ratio 1.2 (1.1–1.3) 1.2 (1.1–1.3) 1.2 (1.1–1.4) 0.865 Width of portal vein,mm 7.1 (3.4–30.1) 7.2 (3.4–30.5) 6.7 (3.3–26.0) 0.375 3-year PVT, n (%) 5-year PVT, n (%) 167(18.3) 217(23.8) 121(19.0) 151(23.7) 46(16.7) 66(24.0) 0.202 0.172 CTP, Child-Turcotte-Pugh; MELD, model of end-stage liver disease; 3.2 Construction of ANN model Table 2 presents the results acquired from the Cox regression analysis. The analysis demonstrated significant associations between several factors and the occurrence of PVT. Age (HR = 1.045, 95% CI 1.029–1.061, p < 0.001), gastrointestinal varices with bleeding (HR = 0.767, 95% CI 1.420–3.265, p < 0.001), ALT (HR = -0.002, 95% CI 0.996–0.999, p = 0.004), ALB (HR = -0.029, 95% CI 0.947–0.997, p = 0.028), NLR (HR = 0.286, 95% CI 1.262–1.403, p < 0.001), PLT (HR = -0.022, 95% CI 0.972–0.985, p < 0.001), International normalized ratio (HR = 0.559, 95% CI 0.936–3.628, p = 0.008), and Width of portal vein (HR = 0.002, 95% CI 0.996-1.000, p = 0.003) were all identified as significantly associated with PVT occurrence in the training group. Table 2 Factors associated with prediction incidence of PVT Variables Univariate analysis β HR (95% CI) P value Age (yr) 0.044 1.045 (1.029,1.061) < 0.001 Sex(male) 0.267 1.306 (0.873,1.954) 0.194 Smoking 0.081 0.922 (0.605,1.405) 0.706 Alcohol consumption 0.161 1,175 (0.767,1.800) 0.458 Diabetes 0.255 1.291 (0.807,2.064) 0.287 Ascites 0.660 1.935 (1.354,2.766) < 0.001 Hepatic encephalopathy 0.255 1.291 (0.807,2.064) 0.287 Gastrointestinal varices with bleeding 0.767 2.153 (1.420,3.265) < 0.001 Alanine aminotransferase (U/L) -0.002 0.998 (0.996,0.999) 0.004 Aspartate aminotransferase (U/L) -0.002 0.998 (0.996,0.999) 0.323 Total bilirubin (mg/dl) -0.003 0.997 (0.992,1.001) 0.114 Albumin (g/L) -0.029 0.972 (0.947,0.997) 0.028 gamma-glutamyl transpeptidase (U/L) 0.001 1.001 (0.998,1.003) 0.637 White blood cell count (×109/L) -0.007 0.993 (0.913,1.079) 0.861 NLR 0.286 1.331 (1.262,1.403) < 0.001 Platelets (×109/L) -0.022 0.979 (0.972,0.985) < 0.001 Creatinine (µmol/L) 0.001 1.001 (0.995,1.007) 0.758 International normalized ratio 0.559 1.749 (0.936,3.268) 0.080 Width of portal vein -0.002 0.998 (0.996,1.000) 0.063 OR, odds ratio; 95% CI, 95% confidence interval; MELD model of end-stage liver disease These factors were also incorporated in constructing the artificial neural network (ANN) model available at https://lixuan.me/annmodel/myg-v4/ . The multilayer perceptron (MLP) is a frequently used structure for ANN, consisting of input, hidden, and output layers. The input layer comprises clinical and biochemical parameters, while the output layer entails corresponding prognosis outcomes ( 22 ). The ANN model for predicting the 3/5-year risk of PVT development in patients with cirrhosis can be accessed at https://houyixin.math.ink/PVTR/index.html . In the model, neurons are interconnected through weighted links, resulting in a total of 8 input neurons and two output neurons. To optimize the MLP's performance, we incorporated four hidden layers after rigorous debugging and testing. Application of the ANN model for risk stratification We categorized all patients into three different groups based on the upper quartiles and lower quartiles of the ANNs model scores: Strata 1, representing low risk, Strata 2, medium risk, and Strata 3 representing high risk. In the training cohort, when comparing Strata3 to Strata 1 as reference, the hazard ratios (HRs) for Strata 3 were 0.8 (95%CI 29.11–86.82) (P < 0.0001). Similarly, in the validation cohort, we observed noticeable survival differences across all stratifications. The ANNs model was successful in accurately distinguishing between patients based on their different risks, whether it was in the training cohort or validation cohort. In terms of the validation cohort, the positive predictive value for low risk was 26.2% (95%CI 25.0-27.4), while the negative predictive value was 98.7% (95%CI 95.2–99.7). On the other hand, the positive predictive value for high risk was 54.7% (95%CI 48.6–60.7), with a negative predictive value of 91.6% (95%CI 89.4–93.4). In another instance of the validation cohort, the positive predictive value for low risk was 20.9% (95%CI 19.6–22.2), and the negative predictive value was 100% (-). Likewise, the positive predictive value for high risk was 41.5% (95%CI 32.8–50.8), and the negative predictive value was 91.9% (95%CI 88.6–94.3) (Table 3 ). Table 3 Positive predictive and negative predictive values. Cohort Models 3-year risk of PVT 5-year risk of PVT Positive(%)(95%CI) Negative(%) (95%CI) Positive(%)(95%CI) Negative(%) (95%CI) Training ANN(low) ANN (high) 26.2(25.0-27.4) 54.7(48.6–60.7) 98.7(95.2–99.7) 98.6(89.4–99.4) 23.2(21.0-28.4) 52.7(49.6–60.7) 97.7(95.2–99.7) 98.6(88.4–99.5) Validation ANN(low) 20.9 (19.6–22.2) 100(-) 19.9 (19.6–24.2) 97.9 (89.6–98.3) ANN(high) 41.5(32.8–50.8) 91.9 (88.6–94.3) 31.5(28.8–51.8) 98.9 (88.6–99.3) From there, we inputted the patient information into the Ann model, enabling us to determine the patient's risk of developing PVT over a span of 3–5 years, and subsequently classify the patient's risk accordingly (Fig. 2 ). Within the training set, 159 patients (25.0%) were part of the low risk group, 318 patients (49.9%) were in the intermediate risk group, and 160 patients (25.1%) were classified as high risk. In the training cohort, we noticed that the predicted cumulative PVT incidence aligns with the observed Kaplan-Meier PVT incidence in the low and medium risk groups (Fig. 3 -AB). Similarly, within the validation set, the plots indicate a remarkable consistency between the observed and predicted cumulative incidence (Fig. 3 -CD) Discrimination and calibration of the ANN model In the training cohort, the occurrence of PVT was accurately predicted by the ANN model, with an AUROC of 0.9718 (95%CI 0.9588–0.9847) and C-index values of 0.9543 (95% CI 0.812–0.974) (Table 4 ). The performance of the ANN model, as measured by AUROC and C-index values, was significantly superior to that of the MELD and CTP models (p < 0.001). Similarly, in the validation cohort, the ANN model demonstrated excellent predictive capability for PVT, with an AUROC of 0.9633 (95% CI 0.9369–0.9897) and a C-index value of 0.9309 (95% CI 0.853–0.965). These values were significantly higher than those of the MELD and CTP models (p < 0.001). (Fig. 4 ). Table 4 Comparison of performance and discriminative ability among the current model and other models. Cohort Models 3-year risk of PVT 5-year risk of PVT AUROC (95%CI) C-index (95%CI) AUROC (95%CI) C-index (95%CI) Training ANN 0.9718(0.9588–0.9847) 0.9543(0.812–0.974) 0.9749(0.9548–0.9915) 0.9580(0.9467–0.9914) Validation ANN 0.9633 (0.9369–0.9897) 0.9409 (0.853–0.965) 0.9727(0.9445–1.0009) 0.9480(0.9487–0.9824) Indecision curve analysis further confirmed the superiority of the ANN model over the MELD and CTP models in both the training and validation cohorts (Fig. 4 ). The calibration curves also indicated good agreement between the predicted probability of PVT-free by the ANN model and the observed probability over 3/5 years in both the training (Fig. 5AB) and validation cohorts (Fig. 5CD). These findings suggest that the ANN model exhibits better clinical practicability compared to other models. Discussion Portal vein thrombosis (PVT) is a frequently occurring complication in individuals with advanced cirrhosis. The presence of PVT exacerbates complications such as stubborn ascites and upper gastrointestinal bleeding by diminishing hepatic blood flow and elevating portal pressure. PVT serves as an indication of advanced disease and directly influences a patient's prognosis. However, detecting PVT in patients with cirrhosis and its treatment can be challenging due to the lack of apparent clinical manifestations during the early stages. Consequently, severe complications often arise, adversely impacting the prognosis of individuals with liver cirrhosis. Thus, early identification of high-risk factors contributing to PVT formation is crucial for preventing its development. This approach significantly delays the progression of liver cirrhosis and enhances the survival rate of patients. The current study employed a machine learning-based artificial neural network (ANN) prediction model specifically designed to assess the probability of PVT occurrence within a 3–5 year timeframe for individual patients. Our research indicates that various factors, such as presence of gastrointestinal varices accompanied by bleeding, ALB levels, PLT levels, International normalized ratio, and Width of portal vein, are closely linked to the development of PVT. Previous investigations have also proposed that decompensation and a low platelet count are potential risk factors for PVT( 31 ). Nevertheless, there has been some controversy surrounding the reproducibility of Width of portal vein as a determining factor for PVT development( 32 ). Our study offers compelling evidence that an accurate measurement of Width of portal vein can serve as a valuable predictive tool for PVT development. Furthermore, our findings imply that individuals with hepatitis B cirrhosis who have encountered upper gastrointestinal bleeding are at an elevated risk for PVT. This could be attributed to the administration of hemostatic treatment after bleeding incidents, which could induce a hypercoagulable state and raise the likelihood of PVT. The performance of the model based on artificial neural network (ANN) in predicting the occurrence of portal vein thrombosis (PVT) at 3/5 years was exceptional. This was confirmed by the significantly high area under the curve (AUC) value of 0.956 obtained from training and calibration curves. In contrast, other models like MELD and CTP demonstrated lower AUC values. The remarkable predictive ability of our ANN model was particularly evident in patients diagnosed with cirrhosis. One advantage offered by ANN models is their capacity to acquire knowledge from data and optimize prediction accuracy by adjusting the connections among variables. In contrast to traditional logistic regression or Cox regression models, ANN models are non-linear and iteratively train the factors relevant to the outcome. This characteristic enables them to achieve a greater level of prediction accuracy ( 33 , 34 , 35 ). It is important to acknowledge several limitations of this study that need to be considered. Firstly, it is crucial to recognize that this study has a retrospective design, which introduces a certain degree of bias in participant selection. Secondly, it was not feasible to compare the baseline data of the external validation group, thus necessitating further studies to supplement this aspect. Moreover, this study did not collect relevant indicators such as portal vein blood flow velocity, splenic vein diameter, spleen thickness, thromboelastography, among others. Incorporating these indicators in future studies has the potential to enhance the accuracy of the prediction model. The well-established deep neural network model showcases robust capability in predicting concurrent PVT in patients diagnosed with cirrhosis. It serves as a user-friendly and easily implementable tool for clinical application. Regular evaluation and examination of relevant indicators throughout the management of cirrhosis patients are crucial. This approach enables early identification of high-risk PVT patients and facilitates appropriate clinical decision-making to improve their prognosis Summary This study employed an ANN model to construct a predictive tool for estimating the 3/5-year risk of PVT development in patients with cirrhosis. The ANN model demonstrated promising individualized prediction performance, thereby offering valuable assessment of PVT risk in clinical settings for patients with cirrhosis. Declarations Ethical Approval and consent to participate The study was approved by the Ethics Committee of Beijing Ditan Hospital, Capital Medical University. Written informed consent was obtained from each patient. All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008. Consent to Publication-NA Data Availability statement The dataset used and analysed during the study can be available from the corresponding author on reasonable request Conflict of Interest The authors declare that they have no conflicts of interest with regard to the publication of this research report. Funding This work was supported by Beijing Hospitals Authority Youth Programme（QMl220201802），the Beijing Traditional Chinese medicine science and Technology Development Fund Project (No. Qn-2020-25), Application of Clinical Features of Capital City of Science and Technology commission (z181100001718052)，Beijing Municipal Administration of Hospitals Incubating Program(PZ2024034) Acknowledgments We gratefully recognize the patients who participated in this study. We thank for Lihua Yu her help with the data collection. Author Contributions YH, PM and YZ designed the study and interpreted the results. XL,TW,HY and XJ collected the data and carried out analysis. All authors read and approved the final manuscript References Ambrosino P, Tarantino L, Di Minno G, Paternoster M, Graziano V, PetittoM, et al. The risk of venous thromboembolism in patients with cirrhosis. Thromb Haemost. 2017;117:139–48. Francoz C, Valla D, Durand F. Portal vein thrombosis, cirrhosis, and liver transplantation. J Hepatol. 2012;57(1):203–12. Nery F, Chevret S, Condat B, de Raucourt E, Boudaoud L, Rautou PE, etal. Causes and consequences of portal vein thrombosis in 1,243 patients with cirrhosis: Results of a longitudinal study. Hepatology.2015;61(2):660–7. Zocco MA, Di Stasio E, De Cristofaro R, Novi M, Ainora ME, Ponziani F,et al. Thrombotic risk factors in patients with liver cirrhosis: Correlation with MELD scoring system and portal vein thrombosis development. J Hepatol. 2009;51(4):682–9. Maruyama H, Okugawa H, Takahashi M, Yokosuka O. De novo Portal Vein Thrombosis in Virus-Related Cirrhosis: Predictive Factors and Long-Term Outcomes. Am J Gastroenterol. 2013;108(4):568–74. Noronha-Ferreira C, Marinho RT, Cortez-Pinto H, Ferreira P, Dias MS,Vasconcelos M, et al. Incidence, predictive factors and clinical significance of development of portal vein thrombosis in cirrhosis: A prospective study. Liver Int. 2019;39(8):1459–67. Nery F, Correia S, Macedo C, Gandara J, Lopes V, Valadares D, et al.Nonselective beta-blockers and the risk of portal vein thrombosis in patients with cirrhosis: results of a prospective longitudinal study. Aliment Pharmacol Ther. 2019;49(5):582–8. Gaballa D, Bezinover D, Kadry Z, Eyster E, Wang M, Northup PG, et al.Development of a Model to Predict Portal Vein Thrombosis in Liver Transplant Candidates: The Portal Vein Thrombosis Risk Index. Liver Transplant. 2019;25(12):1747–55. Stine JG, Wang J, Shah PM, Argo CK, Intagliata N, Uflacker A, et al.Decreased portal vein velocity is predictive of the development of portalvein thrombosis: A matched case-control study. Liver Int. 2018;38(1):94–101. Francoz C, Belghiti J, Vilgrain V, Sommacale D, Paradis V, Condat B, etal. Splanchnic vein thrombosis in candidates for liver transplantation:usefulness of screening and anticoagulation. Gut. 2005;54(5):691–7. Basili S, Carnevale R, Nocella C, Bartimoccia S, Raparelli V, et al. Serum Albumin Is Inversely Associated With Portal Vein Thrombosis in Cirrhosis. Hepatol Commun. 2019;3(4):504–12. Giannitrapani L, Granà W, Licata A, Schiavone C, Montalto G, Soresi M. Nontumorous portal vein thrombosis in liver cirrhosis: Possible role of β-blockers. Med Princ Pract. 2018;27(5):466–71. Yerdel M a, Gunson B, Mirza D, Karayalçin K, Olliff S, Buckels J, et al.Portal vein thrombosis in adults undergoing liver transplantation: riskfactors, screening, management, and outcome. Transplantation.2000;69(9):1873–81. Xu X, Guo X, De Stefano V, Silva-Junior G, Goyal H, Bai Z, et al.Nonselective beta-blockers and development of portal vein thrombosis in liver cirrhosis: a systematic review and meta-analysis. Hepatology International.2019;13(4):468–81. Lisman T, Porte RJ. Rebalanced hemostasis in patients with liverdisease: Evidence and clinical consequences. Blood. 2010;116(6):878–85 Kalambokis GN, Oikonomou A, Christou L, Baltayiannis G. High von Willebrand Factor Antigen Levels and Procoagulant Imbalance May BeInvolved in Both Increasing Severity of Cirrhosis and Portal VeinThrombosis. 2016;64(4):1383–5. La Mura V, Tripodi A, Tosetti G, Cavallaro F, Chantarangkul V, Colombo M, et al. Resistance to thrombomodulin is associated with de novo portal vein thrombosis and low survival in patients with cirrhosis. Liver Int. 2016;36:1–9. Martinelli I, Primignani M, Aghemo A, Reati R, Bucciarelli P, Fabris F, etal. High levels of factor VIII and risk of extra-hepatic portal veinobstruction. J Hepatol. 2009;50(5):916–22. Lancellotti S, Basso M, Veca V, Sacco M, Riccardi L, Pompili M, et al.Presence of portal vein thrombosis in liver cirrhosis is strongly associatedwith low levels of ADAMTS-13: a pilot study. Intern Emerg Med.2016;11(7):959–67. Ma SD, Wang J, Bezinover D, Kadry Z, Northup PG, Stine JG. Inherited thrombophilia and portal vein thrombosis in cirrhosis: A systematic review and meta-analysis. Res Pract Thromb Haemost. 2019;3(4):658–67. Mangia A, Villani MR, Cappucci G, Santoro R, Ricciardi R, Facciorusso D,et al. Causes of portal venous thrombosis in cirrhotic patients: The role ofgenetic and acquired factors. Eur J Gastroenterol Hepatol.2005;17(7):745–51. Engelmann B, Massberg S. Thrombosis as an intravascular effector ofinnate immunity. Nat Rev Immunol.2013;13(1):34–45. Jiménez-Alcázar M, Kim N, Fuchs TA. Circulating Extracellular DNA:Cause or Consequence of Thrombosis Semin Thromb Hemost.2017;43(6):553–61. Berzigotti A, Piscaglia F. Ultrasound in Portal Hypertension – Part 1.Ultraschall Med. 2011;32:548–71. La Mura V, Reverter JC, Flores-Arroyo A, Raffa S, Reverter E, Seijo S, etal. Von Willebrand factor levels predict clinical outcome in patients with cirrhosis and portal hypertension. Gut. 2011;60(8):1133–8. Raffa S, Reverter JC, Seijo S, Tassies D, Abraldes JG, Bosch J, et al. Hypercoagulability in patients with chronic noncirrhotic portal vein thrombosis. Clin Gastroenterol Hepatol. 2012;10(1):72–8. Ariëns RAS, Kohler HP, Mansfield MW, Grant PJ. Subunit antigen and activity levels of blood coagulation factor XIII in healthy individuals:Relation to sex, age, smoking, and hypertension. Arterioscler ThrombVasc Biol. 1999;19(8):2012–6. Hemker HC, Giesen P, Al Dieri R, Regnault V, De Smedt E, WagenvoordR, et al. Calibrated automated thrombin generation measurement in clotting plasma. Pathophysiol Haemost Thromb. 2003;33(1):4–15. Martínez-Zamora MA, Tàssies D, Creus M, Reverter JC, Puerto B,Monteagudo J, et al. Higher levels of procoagulant microparticles in women with recurrent miscarriage are not associated with antiphospholipid antibodies. Hum Reprod. 2016;31(1):46–52. Von Meijenfeldt FA, Burlage LC, Bos S, Adelmeijer J, Porte RJ, Lisman T. Elevated Plasma Levels of Cell-Free DNA During Liver Transplantation Are Associated With Activation of Coagulation. Liver Transplant.2018;24(12):1716–25. Lisman T, De Groot PG, Meijers JCM, Rosendaal FR. Reduced plasma fibrinolytic potential is a risk factor for venous thrombosis. Blood. 2005;105(3):1102–5. Aleman MM, Byrnes JR, Wang JG, Tran R, Lam WA, Paola J Di, et al.Factor XIII activity mediates red blood cell retention in venous thrombi. J Clin Invest. 2014;124(8):3590–600. Lisman T, Moschatsis S, Adelmeijer J, Karel Nieuwenhuis H, De Groot PG. Recombinant factor VIIa enhances deposition of platelets with congenital or acquired αIIbβ3 deficiency to endothelial cell matrix and collagen under conditions of flow via tissue factor-independent thrombin generation. Blood. 2003;101(5):1864–70. Lisman T, Ariëns RAS. Alterations in Fibrin Structure in Patients with Liver Diseases. Semin Thromb Hemost. 2016;42(4):389–96. Hugenholtz GCG, Macrae F, Adelmeijer J, Dulfer S, Porte RJ, Lisman T,et al. Procoagulant changes in fibrin clot structure in patients withcirrhosis are associated with oxidative modifications of fibrinogen. J Thromb Haemost. 2016;14(5):1054–66 Additional Declarations No competing interests reported. Supplementary Files trainingdataset.xlsx validationdataset.xlsx Cite Share Download PDF Status: Posted Version 1 posted 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-3831874","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":265819398,"identity":"f421c86d-7c24-40ab-ac09-5de900364fbe","order_by":0,"name":"Peipei Meng","email":"","orcid":"","institution":"Center of Integrative Medicine, Beijing Ditan Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Peipei","middleName":"","lastName":"Meng","suffix":""},{"id":265819399,"identity":"54af235d-8928-4277-b5f3-71756a967f9a","order_by":1,"name":"Yang Zhou","email":"","orcid":"","institution":"Center of Integrative Medicine, Beijing Ditan Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Zhou","suffix":""},{"id":265819400,"identity":"7a8ad2e0-68de-4682-ba69-8dd42ec7c4f2","order_by":2,"name":"Xiaoli Liu","email":"","orcid":"","institution":"Center of Integrative Medicine, Beijing Ditan Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xiaoli","middleName":"","lastName":"Liu","suffix":""},{"id":265819401,"identity":"9429135e-8e2d-4323-8aea-3e589c7f32d5","order_by":3,"name":"Tong Wu","email":"","orcid":"","institution":"Center of Integrative Medicine, Beijing Ditan Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Tong","middleName":"","lastName":"Wu","suffix":""},{"id":265819402,"identity":"dbdc5de1-ce17-40b9-9f18-588df019bf3f","order_by":4,"name":"Hao Yu","email":"","orcid":"","institution":"Center of Integrative Medicine, Beijing Ditan Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Hao","middleName":"","lastName":"Yu","suffix":""},{"id":265819403,"identity":"5f9fdd82-7563-4731-aae4-1f14cf23c52f","order_by":5,"name":"Xiaomin Ji","email":"","orcid":"","institution":"Center of Integrative Medicine, Beijing Ditan Hospital, Capital Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xiaomin","middleName":"","lastName":"Ji","suffix":""},{"id":265819404,"identity":"541c3c6a-160e-45dc-a6d6-859dfbd46f81","order_by":6,"name":"Yixin Hou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxklEQVRIiWNgGAWjYLCCBAYJOX72xsaHH0jQYmEs2XO42ViCBHsqEg1upLcJ8BCj1uBG8jGJhzskEhhuPmxjkGCwk9NtIKBFckZaskHiGYk8xtmJbQ8KGJKNzQ4Q0MIvkWP4ILFNophZOrHdQILhQOI2QlrYJPI/HABqSWyTPNgmwUOMFqAtjCBbEnskGInUItnzzNgAqMVYgicRGMgGRPjF4HjyM8mfbXVy9sePP3z4ocJOjqAWdBNIUz4KRsEoGAWjAAcAAEHQP/GQEE+/AAAAAElFTkSuQmCC","orcid":"","institution":"Center of Integrative Medicine, Beijing Ditan Hospital, Capital Medical University","correspondingAuthor":true,"prefix":"","firstName":"Yixin","middleName":"","lastName":"Hou","suffix":""}],"badges":[],"createdAt":"2024-01-03 13:01:50","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3831874/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3831874/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":49438999,"identity":"643815a8-3bd6-44d1-8ee8-5156f401cbad","added_by":"auto","created_at":"2024-01-10 21:41:55","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":33148,"visible":true,"origin":"","legend":"\u003cp\u003eStudy flow diagram. HBV, hepatitis B virus.\u003c/p\u003e","description":"","filename":"Fig1Studyflowdiagram.HBVhepatitisBvirus.png","url":"https://assets-eu.researchsquare.com/files/rs-3831874/v1/b9cc9895d8b036705682d6d3.png"},{"id":49439000,"identity":"cfe9f535-c6b4-48c2-8ac2-35f6bb5ef61a","added_by":"auto","created_at":"2024-01-10 21:41:55","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":231055,"visible":true,"origin":"","legend":"\u003cp\u003eArtificial neural network model page design according to different conditions of patients\u003c/p\u003e","description":"","filename":"Fig2Artificialneuralnetworkmodelpagedesignaccordingtodifferentconditionsofpatients.png","url":"https://assets-eu.researchsquare.com/files/rs-3831874/v1/ceb01eb67f6501d14dabdb28.png"},{"id":49439001,"identity":"8a3e0d84-5336-401c-aae4-74c76212b9db","added_by":"auto","created_at":"2024-01-10 21:41:55","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":152555,"visible":true,"origin":"","legend":"\u003cp\u003eAccording to the ANN model, the patient training and validation sets are divided into three risk layers as follows: \u0026nbsp;high ,medium, and low.\u003c/p\u003e","description":"","filename":"Fig3AccordingtotheANNmodelthepatienttrainingandvalidationsetsaredividedintothreerisklayersasfollowshighmediumandlow.png","url":"https://assets-eu.researchsquare.com/files/rs-3831874/v1/a7e1cb222ce7284a8dd865d2.png"},{"id":49439005,"identity":"345ef777-75c1-48ad-8bbc-255db5cdbd8a","added_by":"auto","created_at":"2024-01-10 21:41:55","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":102359,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted versus observed cumulative incidence of \u0026nbsp;PVT based on the predictive model.\u003c/p\u003e","description":"","filename":"Fig4PredictedversusobservedcumulativeincidenceofPVTbasedonthepredictivemodel..png","url":"https://assets-eu.researchsquare.com/files/rs-3831874/v1/1cf9745132cf4ad760a7fe70.png"},{"id":49439747,"identity":"50e319c4-ebbd-4084-b549-acf99d79fc30","added_by":"auto","created_at":"2024-01-10 21:49:55","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":85811,"visible":true,"origin":"","legend":"\u003cp\u003eThe cumulative probabilities of PVT of our model at 3/5 years in the training (AB) and validation (CD) data sets.\u003c/p\u003e","description":"","filename":"Fig5ThecumulativeprobabilitiesofPVTofourmodelat35yearsinthetrainingABandvalidationCDdatasets.png","url":"https://assets-eu.researchsquare.com/files/rs-3831874/v1/8bdb57ef67bfc20c2726b65f.png"},{"id":57691271,"identity":"79423866-6fa0-4ff6-9363-5351c3a08a1c","added_by":"auto","created_at":"2024-06-04 11:22:53","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1157352,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3831874/v1/0b026566-9d26-4caa-a93f-44b50139e0ad.pdf"},{"id":49439004,"identity":"667e8d12-6717-4f43-b117-e8f96531c45f","added_by":"auto","created_at":"2024-01-10 21:41:55","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":253152,"visible":true,"origin":"","legend":"","description":"","filename":"trainingdataset.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3831874/v1/073879f8fbfdc36c856176de.xlsx"},{"id":49439003,"identity":"0a3c32bb-656f-4218-9edf-ddc6eeee3d06","added_by":"auto","created_at":"2024-01-10 21:41:55","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":115541,"visible":true,"origin":"","legend":"","description":"","filename":"validationdataset.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3831874/v1/7165b3e8117a2780e368dc2f.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Machine learning-based model used for predicting Portal vein thrombosis risk for patients with cirrhosis","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePortal Vein Thrombosis (PVT) is a type of thrombosis that occurs in the main portal vein and its branches, which can result in partial or complete blockage of the blood vessel. Traditionally, it has been believed that individuals with cirrhosis of the liver experience clotting factor synthesis disorders and thrombocytopenia, making them more prone to bleeding. However, recent research suggests that the anticoagulant and procoagulant systems in the liver of individuals with cirrhosis are in a dynamic and unstable state, leading to stagnation of blood flow in the portal vein and dysfunction of the blood vessel lining. This can increase the risk of both bleeding and thrombosis (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). In fact, non-tumoral PVT is relatively common in patients with cirrhosis, with an estimated annual incidence ranging from 4.6\u0026ndash;26% (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e), and higher in those with more advanced liver disease.\u003c/p\u003e \u003cp\u003eCurrently, most studies on PVT analyze influencing factors using retrospective data, and there is no reliable model for early prediction of PVT occurrence. As a result, identifying high-risk individuals for early prevention is challenging in clinical practice (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Some studies have suggested that a portal blood flow velocity (PBFV) below 15 cm/sec (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e) is a major risk factor for PVT development. Other risk factors include the severity of liver disease and the presence of portal hypertension, such as low platelet count (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e), low albumin levels (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e), large esophageal varices (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e), previous sclerotherapy (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e), previous liver decompensation (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e), and the presence of large portosystemic collaterals (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). More recently, it has been proposed that the Model for End-Stage Liver Disease (MELD) and Child-Pugh-Turcotte (CTP) scores can help predict the development of PVT (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). However, these studies did not adequately consider the potential influence of other recognized risk factors for PVT\u003c/p\u003e \u003cp\u003eArtificial Neural Networks (ANNs) are a form of machine learning that mimics the information processing of brain neurons (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). These mathematical models have found extensive applications in medical decision-making by analyzing linear, logistic, and nonlinear complex relationships (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Through training, ANNs optimize the factors associated with the outcome, resulting in highly accurate prediction models. Therefore, this study aims to employ ANNs in constructing an advanced warning model for accurate identification of high-risk groups prone to disease progression, specifically predicting the PVT\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec4\" class=\"Section3\"\u003e \u003ch2\u003e2.1 Patients\u003c/h2\u003e \u003cp\u003eA total of 1505 patients diagnosed with cirrhosis were retrospectively enrolled at the Beijing Ditan Hospital of Capital Medical University in Beijing, China, from January 2011 to January 2016. Eligibility criteria included being a first-time diagnosis of cirrhosis, age between 18 and 80 years, and confirmation of cirrhosis through liver biopsy and/or compatible clinical, laboratory, and imaging data. Patients who died within a 3-year or 5-year period or were lost to follow-up were excluded. Exclusion criteria encompassed individuals with known hepatocellular carcinoma (HCC), pregnant women, those who have undergone previous orthotopic liver transplantation (OLT), unwillingness to provide informed consent, and use of anticoagulation therapy or prior surgical or transjugular intrahepatic portosystemic shunt (TIPS) procedures. To ensure a representative sample, 986 patients were randomly assigned, with 70% (n\u0026thinsp;=\u0026thinsp;685) allocated to the training cohort and the remaining 30% (n\u0026thinsp;=\u0026thinsp;301) assigned to the validation cohort (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). This study received ethical approval from the Ethics Committee of Beijing Ditan Hospita\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Clinical definition and follow-up\u003c/h2\u003e \u003cp\u003eCompensatory cirrhosis was determined through the following methods: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) On biopsy, the presence of pathological findings indicating F4 stage cirrhosis; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) During endoscopy, the presence of esophageal varices and exclusion of noncirrhotic portal hypertension; (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) In the absence of histological evidence and endoscopic findings, at least two out of three criteria should be met:①Imaging techniques such as ultrasonography, computed tomography, or magnetic resonance imaging show changes in liver morphology, such as nodules in the liver tissue and uneven texture on the liver surface; ②Platelet count lower than 100\u0026times;10\u003csup\u003e9\u003c/sup\u003e cells/L, without any other identifiable causes; ③Serum albumin lower than 35.0g/L, international normalized ratio higher than 1.3, or prothrombin time prolonged by more than 3 seconds. The diagnosis of decompensated cirrhosis is based on the presence of cirrhosis along with complications related to portal and venous hypertension and/or liver dysfunction. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) The diagnosis requires evidence of cirrhosis; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) Presence of complications associated with portal hypertension such as ascites, bleeding from esophageal and gastric varices, and hepatic encephalopathy (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). PVT (portal vein thrombosis) is diagnosed by the visualization of non-tumoral thrombosis within the portal vein or its branches. Such diagnosis and assessment of its extent are always confirmed through computed tomography or magnetic resonance imaging. PVT is categorized as occlusive if there is a complete absence of blood flow in the vein, and partial if the lumen is only partially occluded while blood flow is still present\u003c/p\u003e \u003cp\u003eThe starting point of this research was the initial cirrhosis diagnosis at the hospital. The conclusion of this study focused on the recent identification of PVT within a one-year span and the subsequent follow-up of five years. The collection of clinical data encompassed various categories such as demographics (age, gender), complications (bleeding gastrointestinal varices, ascites, hepatic encephalopathy [HE]), biochemical indicators (alanine aminotransferase [ALT], aspartate aminotransferase [AST], total bilirubin [TBIL], serum albumin [ALB], ɣ-glutamyl transpeptidase [GGT], white blood cell count [WBC], neutrophil count [NC], lymphocyte count [LC], platelet count [PLT], creatinine [CREA], prothrombin time [PT], international normalized ratio [INR], Alpha-fetoprotein [AFP]), and routine laboratory tests which involved computed tomography or magnetic resonance imaging, conducted every 3\u0026ndash;6 months\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Construction of ANN\u003c/h2\u003e \u003cp\u003eThe artificial neural network (ANN) is composed of complex and interconnected processing units called neurons. These neurons are connected through weighted connections and organized into an input layer, output layer, and one or more hidden layers (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). Some advantages of using an ANN include its ability to self-learn, self-adapt, and perform inference processes. The ANN learns by examining examples and adjusting the weights of the connections between neurons to establish a relationship between input and output. When applied to data, the input is passed through the layers of neurons until an output is generated. Following this output, a process of self-adaptation occurs. The produced output is compared to the desired output, and if there is a difference, an error signal is generated. This error signal is then used in a back propagation (BP) method to modify the weights of the connections between neurons. This modification aims to minimize the overall error of the network. Throughout the learning process, the error between the produced outputs and the desired outputs gradually decreases until it reaches a minimum, indicating convergence of the network. Once this convergence is achieved, the ANN can perform an inference process. During this process, new input data can be used to generate outputs or predictions based on the knowledge gained during the training process. This allows the ANN to accurately predict outcomes on different data sets (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e)\u003c/p\u003e \u003cp\u003eIn this investigation, the 5-year progression of PVT in the 637 layers of cirrhosis input contained neurons that imported the available data, encompassing a variety of clinical, demographic, and laboratory information. The output layers consisted of neurons that exported the corresponding predictive results. The hidden layers were utilized to facilitate intricate interactions between the input and output neurons. Factors significantly linked to PVT in patients with cirrhosis were used to construct artificial neural networks (ANNs) employing Mathematica 11.1.1 for Microsoft Windows (64-bit), a graphical tool for neural network development. A total of 912 patients were assigned to either a training group (n\u0026thinsp;=\u0026thinsp;637, 70%) or a validation group (n\u0026thinsp;=\u0026thinsp;275, 30%). The backpropagation (BP) algorithm was employed in the learning process of this ANN, which involved assessing the errors between the generated and desired outputs. The connections between neurons were adjusted by modifying the weights to minimize the overall network errors. The learning (training) process would be terminated once the sum of squared errors reached a minimum compared to the cross-validation dataset. Ultimately, the final model provided the development of PVT risks over a 3/5-year period for each patient.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.4Statistical analysis\u003c/h2\u003e \u003cp\u003eThe data were expressed in the form of median (range) or n (%) as applicable. To assess the statistical significance of differences among continuous and categorical variables, we employed Student's t-test (or Mann-Whitney test if appropriate) and chi-squared test (or Fisher's exact test if appropriate). Once the relationship between demographic, biochemical, and clinical variables (inputs) and prognosis (outputs) was determined, we selected the variables with statistically significant differences or important clinical characteristics as the input layers for constructing artificial neural networks (ANNs) to predict the development of portal vein thrombosis (PVT) in patients over a period of 3 years and 5 years. We presented hazard ratios (HR) and their corresponding 95% confidence intervals (CI), along with p values. To evaluate the discriminatory performance, we utilized receiver-operating characteristic (ROC) curves. The area under the ROC curve was computed to generate Harrell's c-index. Furthermore, we compared the performance of the ANN model with that of Model for End-Stage Liver Disease (MELD) in the ROC curves [28,29,30]. The scores for MELD were calculated based on the published scoring formula. To visually assess the agreement between the predicted probability of PVT over 3/5 years by the model and the observed probability, we employed a calibration plot. Additionally, we conducted decision curve analysis (DCA) to compare the clinical net benefits of the new model compared to previous models. For all statistical tests, a p-value of less than 0.05 was considered to indicate a statistically significant difference. We performed the statistical analysis using SPSS 22 (IBM, Armonk, NY, USA) and R version 3.3.2 (R core development team, 2010).\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Baseline characteristics\u003c/h2\u003e \u003cp\u003eBetween 2008 and 2016, we conducted a study involving the participation of 986 patients. These patients were randomly divided into two datasets, namely the training dataset (n\u0026thinsp;=\u0026thinsp;685) and the validation dataset (n\u0026thinsp;=\u0026thinsp;301). Within the three-year period, 121 patients (19.0%) developed PTV in the training group, while 151 patients (23.7%) developed PTV within five years. In the validation group, there were 46 patients (16.7%) who developed PTV within three years and 66 patients (24.0%) who developed PTV within five years. The characteristics of the two groups were comparable as shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. In the training set, a great majority of the participants, specifically 439 individuals (69.0%), were male, and their median age was 52.8 years (interquartile range: 40\u0026ndash;74 years). Importantly, no significant differences in baseline characteristics were observed between the two sets(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\u003eBasic clinical characteristics of patients with HBV-related cirrhosis.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\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 \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\u003eAll patients (n\u0026thinsp;=\u0026thinsp;912)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTraining Cohort\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;637)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eValidation Cohort\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;275)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge, years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e53.0 (41.0\u0026ndash;73.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e52.8.0 (40.0\u0026ndash;74.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e52.9(43.0\u0026ndash;75.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.436\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale sex, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e611 (67.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e439(69.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e172(62.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.320\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmoking, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e210(23.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e149(23.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e61 (22.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.846\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlcohol consumption, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e195(21.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e123 (19.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e62(22.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.579\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e152 (16.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e103(16.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e49(17.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.174\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertension, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e129 (14.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e95 (14.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e34 (12.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.835\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAscites, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e197(21.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e134 (21.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e63 (22.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.485\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEncephalopathy, n (%)\u003c/p\u003e \u003cp\u003eGastrointestinal varices with bleeding, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e41(4.5)\u003c/p\u003e \u003cp\u003e134(14.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e30 (4.7)\u003c/p\u003e \u003cp\u003e88(13.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e11 (4.0)\u003c/p\u003e \u003cp\u003e46(16.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.458\u003c/p\u003e \u003cp\u003e0.348\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCTP score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7.0 (5.0\u0026ndash;10.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7.0 (6.0\u0026ndash;9.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7.0 (5.0\u0026ndash;10.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.626\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMELD score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e10.1 (7.8\u0026ndash;12.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e10.1 (8.0-12.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e10.1 (7.7\u0026ndash;13.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.851\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlanine aminotransferase, U/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e41.3 (26.5-105.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e43.6 (27.1-112.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e37.5 (25.1\u0026ndash;95.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.164\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAspartate aminotransferase, U/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e47.9 (30.8-106.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e48.6 (32.1-107.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e46.7 (29.7\u0026ndash;98.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.282\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal bilirubin, \u0026micro;mol/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e22.4 (14.2\u0026ndash;39.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e22.7 (14.1\u0026ndash;38.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22.3 (14.5\u0026ndash;40.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.805\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlbumin, g/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e33.6 (30.3\u0026ndash;39.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e33.8 (30.4\u0026ndash;40.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e33.2 (30.1\u0026ndash;38.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.258\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGamma-glutamyl transpeptidase, U/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e56.3 (37.2\u0026ndash;97.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e55.3 (36.1\u0026ndash;99.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e57.3 (40.4\u0026ndash;95.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.497\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWhite blood cell count, \u0026times;10\u003csup\u003e9\u003c/sup\u003e/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3.9 (2.8\u0026ndash;5.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3.9 (2.8\u0026ndash;5.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.8 (2.7\u0026ndash;5.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.451\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNeutrophil count, \u0026times;10\u003csup\u003e9\u003c/sup\u003e/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.2 (1.5\u0026ndash;3.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.2 (1.5\u0026ndash;3.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.2 (1.5-3.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.434\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLymphocyte count, \u0026times;10\u003csup\u003e9\u003c/sup\u003e/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.1 (0.8\u0026ndash;1.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.1 (0.8\u0026ndash;1.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.1 (0.7\u0026ndash;1.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.921\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNeutrophil-lymphocyte ratio\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.0 (1.4\u0026ndash;2.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.0 (1.4\u0026ndash;2.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.9 (1.4\u0026ndash;2.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.598\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePlatelets, \u0026times;10\u003csup\u003e9\u003c/sup\u003e/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e87.0 (65.8-118.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e87.0 (65.0-116.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e89.0 (67.0-121.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.199\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCreatinine, \u0026micro;mol/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e66.0 (56.0\u0026ndash;76.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e66.1 (56.1\u0026ndash;76.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e65.0 (56.0-73.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.602\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlood urea nitrogen, mmol/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e5.1(4.0-6.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5.1(4.0-6.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.2 (4.1\u0026ndash;6.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.717\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProthrombin time, s\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e14.3 (12.7\u0026ndash;16.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e14.3 (12.8\u0026ndash;16.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e14.1 (12.5\u0026ndash;16.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.801\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProthrombin activity, %\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e67.0(54.0\u0026ndash;80.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e67.0(54.0\u0026ndash;80.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e67.0(53.0\u0026ndash;81.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.944\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInternational normalized ratio\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.2 (1.1\u0026ndash;1.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.2 (1.1\u0026ndash;1.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.2 (1.1\u0026ndash;1.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.865\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWidth of portal vein,mm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e7.1 (3.4\u0026ndash;30.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7.2 (3.4\u0026ndash;30.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.7 (3.3\u0026ndash;26.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.375\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3-year PVT, n (%)\u003c/p\u003e \u003cp\u003e5-year PVT, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e167(18.3)\u003c/p\u003e \u003cp\u003e217(23.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e121(19.0)\u003c/p\u003e \u003cp\u003e151(23.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e46(16.7)\u003c/p\u003e \u003cp\u003e66(24.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.202\u003c/p\u003e \u003cp\u003e0.172\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eCTP, Child-Turcotte-Pugh; MELD, model of end-stage liver disease;\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\u003e3.2 Construction of ANN model\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e presents the results acquired from the Cox regression analysis. The analysis demonstrated significant associations between several factors and the occurrence of PVT. Age (HR\u0026thinsp;=\u0026thinsp;1.045, 95% CI 1.029\u0026ndash;1.061, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), gastrointestinal varices with bleeding (HR\u0026thinsp;=\u0026thinsp;0.767, 95% CI 1.420\u0026ndash;3.265, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), ALT (HR = -0.002, 95% CI 0.996\u0026ndash;0.999, p\u0026thinsp;=\u0026thinsp;0.004), ALB (HR = -0.029, 95% CI 0.947\u0026ndash;0.997, p\u0026thinsp;=\u0026thinsp;0.028), NLR (HR\u0026thinsp;=\u0026thinsp;0.286, 95% CI 1.262\u0026ndash;1.403, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), PLT (HR = -0.022, 95% CI 0.972\u0026ndash;0.985, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), International normalized ratio (HR\u0026thinsp;=\u0026thinsp;0.559, 95% CI 0.936\u0026ndash;3.628, p\u0026thinsp;=\u0026thinsp;0.008), and Width of portal vein (HR\u0026thinsp;=\u0026thinsp;0.002, 95% CI 0.996-1.000, p\u0026thinsp;=\u0026thinsp;0.003) were all identified as significantly associated with PVT occurrence in the training group.\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\u003eFactors associated with prediction incidence of PVT\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\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=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\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\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eUnivariate analysis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eβ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHR (95% CI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e value\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (yr)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.044\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.045 (1.029,1.061)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\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\u003eSex(male)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.267\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.306 (0.873,1.954)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.194\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmoking\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.081\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.922 (0.605,1.405)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.706\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlcohol consumption\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.161\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1,175 (0.767,1.800)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.458\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.255\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.291 (0.807,2.064)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.287\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAscites\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.660\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.935 (1.354,2.766)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\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\u003eHepatic encephalopathy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.255\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.291 (0.807,2.064)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.287\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGastrointestinal varices with bleeding\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.767\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.153 (1.420,3.265)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\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\u003eAlanine aminotransferase (U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.998 (0.996,0.999)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAspartate aminotransferase (U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.998 (0.996,0.999)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.323\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal bilirubin (mg/dl)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.997 (0.992,1.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.114\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlbumin (g/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.029\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.972 (0.947,0.997)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003egamma-glutamyl transpeptidase (U/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.001 (0.998,1.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.637\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWhite blood cell count (\u0026times;109/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.993 (0.913,1.079)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.861\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNLR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.286\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.331 (1.262,1.403)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\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\u003ePlatelets (\u0026times;109/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.979 (0.972,0.985)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\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\u003eCreatinine (\u0026micro;mol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.001 (0.995,1.007)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.758\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInternational normalized ratio\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.559\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.749 (0.936,3.268)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.080\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWidth of portal vein\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.998 (0.996,1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.063\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eOR, odds ratio; 95% CI, 95% confidence interval; MELD model of end-stage liver disease\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThese factors were also incorporated in constructing the artificial neural network (ANN) model available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://lixuan.me/annmodel/myg-v4/\u003c/span\u003e\u003cspan address=\"https://lixuan.me/annmodel/myg-v4/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. The multilayer perceptron (MLP) is a frequently used structure for ANN, consisting of input, hidden, and output layers. The input layer comprises clinical and biochemical parameters, while the output layer entails corresponding prognosis outcomes (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). The ANN model for predicting the 3/5-year risk of PVT development in patients with cirrhosis can be accessed at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://houyixin.math.ink/PVTR/index.html\u003c/span\u003e\u003cspan address=\"https://houyixin.math.ink/PVTR/index.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. In the model, neurons are interconnected through weighted links, resulting in a total of 8 input neurons and two output neurons. To optimize the MLP's performance, we incorporated four hidden layers after rigorous debugging and testing.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eApplication of the ANN model for risk stratification\u003c/h2\u003e \u003cp\u003eWe categorized all patients into three different groups based on the upper quartiles and lower quartiles of the ANNs model scores: Strata 1, representing low risk, Strata 2, medium risk, and Strata 3 representing high risk. In the training cohort, when comparing Strata3 to Strata 1 as reference, the hazard ratios (HRs) for Strata 3 were 0.8 (95%CI 29.11\u0026ndash;86.82) (P\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). Similarly, in the validation cohort, we observed noticeable survival differences across all stratifications. The ANNs model was successful in accurately distinguishing between patients based on their different risks, whether it was in the training cohort or validation cohort. In terms of the validation cohort, the positive predictive value for low risk was 26.2% (95%CI 25.0-27.4), while the negative predictive value was 98.7% (95%CI 95.2\u0026ndash;99.7). On the other hand, the positive predictive value for high risk was 54.7% (95%CI 48.6\u0026ndash;60.7), with a negative predictive value of 91.6% (95%CI 89.4\u0026ndash;93.4). In another instance of the validation cohort, the positive predictive value for low risk was 20.9% (95%CI 19.6\u0026ndash;22.2), and the negative predictive value was 100% (-). Likewise, the positive predictive value for high risk was 41.5% (95%CI 32.8\u0026ndash;50.8), and the negative predictive value was 91.9% (95%CI 88.6\u0026ndash;94.3) (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\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\u003ePositive predictive and negative predictive values.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"15\"\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 \u003cdiv align=\"left\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c14\" colnum=\"14\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c15\" colnum=\"15\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eCohort\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003eModels\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"5\" nameend=\"c10\" namest=\"c6\"\u003e \u003cp\u003e3-year risk of PVT\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c13\" namest=\"c11\"\u003e \u003cp\u003e5-year risk of PVT\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"1\" nameend=\"c15\" namest=\"c15\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c6\" namest=\"c4\"\u003e \u003cp\u003ePositive(%)(95%CI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c9\" namest=\"c7\"\u003e \u003cp\u003eNegative(%) (95%CI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c11\" namest=\"c10\"\u003e \u003cp\u003ePositive(%)(95%CI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c15\" namest=\"c12\"\u003e \u003cp\u003eNegative(%) (95%CI)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTraining\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eANN(low)\u003c/p\u003e \u003cp\u003eANN (high)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003e26.2(25.0-27.4)\u003c/p\u003e \u003cp\u003e54.7(48.6\u0026ndash;60.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e98.7(95.2\u0026ndash;99.7)\u003c/p\u003e \u003cp\u003e98.6(89.4\u0026ndash;99.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c12\" namest=\"c9\"\u003e \u003cp\u003e23.2(21.0-28.4)\u003c/p\u003e \u003cp\u003e52.7(49.6\u0026ndash;60.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c14\" namest=\"c13\"\u003e \u003cp\u003e97.7(95.2\u0026ndash;99.7)\u003c/p\u003e \u003cp\u003e98.6(88.4\u0026ndash;99.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c15\" namest=\"c15\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eValidation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eANN(low)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003e20.9 (19.6\u0026ndash;22.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e100(-)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c12\" namest=\"c9\"\u003e \u003cp\u003e19.9 (19.6\u0026ndash;24.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c14\" namest=\"c13\"\u003e \u003cp\u003e97.9 (89.6\u0026ndash;98.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c15\" namest=\"c15\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eANN(high)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003e41.5(32.8\u0026ndash;50.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e91.9 (88.6\u0026ndash;94.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c12\" namest=\"c9\"\u003e \u003cp\u003e31.5(28.8\u0026ndash;51.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c14\" namest=\"c13\"\u003e \u003cp\u003e98.9 (88.6\u0026ndash;99.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"1\" nameend=\"c15\" namest=\"c15\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFrom there, we inputted the patient information into the Ann model, enabling us to determine the patient's risk of developing PVT over a span of 3\u0026ndash;5 years, and subsequently classify the patient's risk accordingly (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Within the training set, 159 patients (25.0%) were part of the low risk group, 318 patients (49.9%) were in the intermediate risk group, and 160 patients (25.1%) were classified as high risk. In the training cohort, we noticed that the predicted cumulative PVT incidence aligns with the observed Kaplan-Meier PVT incidence in the low and medium risk groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e-AB). Similarly, within the validation set, the plots indicate a remarkable consistency between the observed and predicted cumulative incidence (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e-CD)\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eDiscrimination and calibration of the ANN model\u003c/h2\u003e \u003cp\u003eIn the training cohort, the occurrence of PVT was accurately predicted by the ANN model, with an AUROC of 0.9718 (95%CI 0.9588\u0026ndash;0.9847) and C-index values of 0.9543 (95% CI 0.812\u0026ndash;0.974) (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The performance of the ANN model, as measured by AUROC and C-index values, was significantly superior to that of the MELD and CTP models (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Similarly, in the validation cohort, the ANN model demonstrated excellent predictive capability for PVT, with an AUROC of 0.9633 (95% CI 0.9369\u0026ndash;0.9897) and a C-index value of 0.9309 (95% CI 0.853\u0026ndash;0.965). These values were significantly higher than those of the MELD and CTP models (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison of performance and discriminative ability among the current model and other models.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCohort\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eModels\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003e3-year risk of PVT\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c9\" namest=\"c8\"\u003e \u003cp\u003e5-year risk of PVT\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eAUROC (95%CI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003eC-index (95%CI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003eAUROC (95%CI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003eC-index (95%CI)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTraining\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eANN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e0.9718(0.9588\u0026ndash;0.9847)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e0.9543(0.812\u0026ndash;0.974)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003e0.9749(0.9548\u0026ndash;0.9915)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.9580(0.9467\u0026ndash;0.9914)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eValidation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eANN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e0.9633 (0.9369\u0026ndash;0.9897)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003e0.9409 (0.853\u0026ndash;0.965)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003e0.9727(0.9445\u0026ndash;1.0009)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e0.9480(0.9487\u0026ndash;0.9824)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIndecision curve analysis further confirmed the superiority of the ANN model over the MELD and CTP models in both the training and validation cohorts (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The calibration curves also indicated good agreement between the predicted probability of PVT-free by the ANN model and the observed probability over 3/5 years in both the training (Fig.\u0026nbsp;5AB) and validation cohorts (Fig.\u0026nbsp;5CD). These findings suggest that the ANN model exhibits better clinical practicability compared to other models.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003ePortal vein thrombosis (PVT) is a frequently occurring complication in individuals with advanced cirrhosis. The presence of PVT exacerbates complications such as stubborn ascites and upper gastrointestinal bleeding by diminishing hepatic blood flow and elevating portal pressure. PVT serves as an indication of advanced disease and directly influences a patient's prognosis. However, detecting PVT in patients with cirrhosis and its treatment can be challenging due to the lack of apparent clinical manifestations during the early stages. Consequently, severe complications often arise, adversely impacting the prognosis of individuals with liver cirrhosis. Thus, early identification of high-risk factors contributing to PVT formation is crucial for preventing its development. This approach significantly delays the progression of liver cirrhosis and enhances the survival rate of patients. The current study employed a machine learning-based artificial neural network (ANN) prediction model specifically designed to assess the probability of PVT occurrence within a 3–5 year timeframe for individual patients.\u003c/p\u003e \u003cp\u003eOur research indicates that various factors, such as presence of gastrointestinal varices accompanied by bleeding, ALB levels, PLT levels, International normalized ratio, and Width of portal vein, are closely linked to the development of PVT. Previous investigations have also proposed that decompensation and a low platelet count are potential risk factors for PVT(\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). Nevertheless, there has been some controversy surrounding the reproducibility of Width of portal vein as a determining factor for PVT development(\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). Our study offers compelling evidence that an accurate measurement of Width of portal vein can serve as a valuable predictive tool for PVT development. Furthermore, our findings imply that individuals with hepatitis B cirrhosis who have encountered upper gastrointestinal bleeding are at an elevated risk for PVT. This could be attributed to the administration of hemostatic treatment after bleeding incidents, which could induce a hypercoagulable state and raise the likelihood of PVT.\u003c/p\u003e \u003cp\u003eThe performance of the model based on artificial neural network (ANN) in predicting the occurrence of portal vein thrombosis (PVT) at 3/5 years was exceptional. This was confirmed by the significantly high area under the curve (AUC) value of 0.956 obtained from training and calibration curves. In contrast, other models like MELD and CTP demonstrated lower AUC values. The remarkable predictive ability of our ANN model was particularly evident in patients diagnosed with cirrhosis. One advantage offered by ANN models is their capacity to acquire knowledge from data and optimize prediction accuracy by adjusting the connections among variables. In contrast to traditional logistic regression or Cox regression models, ANN models are non-linear and iteratively train the factors relevant to the outcome. This characteristic enables them to achieve a greater level of prediction accuracy (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIt is important to acknowledge several limitations of this study that need to be considered. Firstly, it is crucial to recognize that this study has a retrospective design, which introduces a certain degree of bias in participant selection. Secondly, it was not feasible to compare the baseline data of the external validation group, thus necessitating further studies to supplement this aspect. Moreover, this study did not collect relevant indicators such as portal vein blood flow velocity, splenic vein diameter, spleen thickness, thromboelastography, among others. Incorporating these indicators in future studies has the potential to enhance the accuracy of the prediction model. The well-established deep neural network model showcases robust capability in predicting concurrent PVT in patients diagnosed with cirrhosis.\u003c/p\u003e \u003cp\u003eIt serves as a user-friendly and easily implementable tool for clinical application. Regular evaluation and examination of relevant indicators throughout the management of cirrhosis patients are crucial. This approach enables early identification of high-risk PVT patients and facilitates appropriate clinical decision-making to improve their prognosis\u003c/p\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003cp\u003e\u003c/p\u003e \u003c/div\u003e"},{"header":"Summary","content":"\u003cp\u003eThis study employed an ANN model to construct a predictive tool for estimating the 3/5-year risk of PVT development in patients with cirrhosis. The ANN model demonstrated promising individualized prediction performance, thereby offering valuable assessment of PVT risk in clinical settings for patients with cirrhosis.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthical Approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was approved by the Ethics Committee of Beijing Ditan Hospital, Capital Medical University. Written informed consent was obtained from each patient. All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Publication-NA\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe dataset used and analysed during the study can be available from the corresponding author on reasonable request\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no conflicts of interest with regard to the publication of this research report.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by Beijing Hospitals Authority Youth Programme（QMl220201802），the Beijing Traditional Chinese medicine science and Technology Development Fund Project (No. Qn-2020-25), Application of Clinical Features of Capital City of Science and Technology commission (z181100001718052)，Beijing Municipal Administration of Hospitals \u0026nbsp;Incubating Program(PZ2024034)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe gratefully recognize the patients who participated in this study. We thank for Lihua Yu her help with the data collection.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYH, PM and YZ designed the study and interpreted the results. XL,TW,HY and XJ collected the data and carried out analysis. All authors read and approved the final manuscript\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAmbrosino P, Tarantino L, Di Minno G, Paternoster M, Graziano V, PetittoM, et al. The risk of venous thromboembolism in patients with cirrhosis. Thromb Haemost. 2017;117:139\u0026ndash;48.\u003c/li\u003e\n\u003cli\u003eFrancoz C, Valla D, Durand F. Portal vein thrombosis, cirrhosis, and liver transplantation. J Hepatol. 2012;57(1):203\u0026ndash;12.\u003c/li\u003e\n\u003cli\u003eNery F, Chevret S, Condat B, de Raucourt E, Boudaoud L, Rautou PE, etal. Causes and consequences of portal vein thrombosis in 1,243 patients with cirrhosis: Results of a longitudinal study. Hepatology.2015;61(2):660\u0026ndash;7.\u003c/li\u003e\n\u003cli\u003eZocco MA, Di Stasio E, De Cristofaro R, Novi M, Ainora ME, Ponziani F,et al. Thrombotic risk factors in patients with liver cirrhosis: Correlation with MELD scoring system and portal vein thrombosis development. J Hepatol. 2009;51(4):682\u0026ndash;9.\u003c/li\u003e\n\u003cli\u003eMaruyama H, Okugawa H, Takahashi M, Yokosuka O. De novo Portal Vein Thrombosis in Virus-Related Cirrhosis: Predictive Factors and Long-Term Outcomes. Am J Gastroenterol. 2013;108(4):568\u0026ndash;74.\u003c/li\u003e\n\u003cli\u003eNoronha-Ferreira C, Marinho RT, Cortez-Pinto H, Ferreira P, Dias MS,Vasconcelos M, et al. Incidence, predictive factors and clinical significance of development of portal vein thrombosis in cirrhosis: A prospective study. Liver Int. 2019;39(8):1459\u0026ndash;67.\u003c/li\u003e\n\u003cli\u003eNery F, Correia S, Macedo C, Gandara J, Lopes V, Valadares D, et al.Nonselective beta-blockers and the risk of portal vein thrombosis in patients with cirrhosis: results of a prospective longitudinal study. Aliment Pharmacol Ther. 2019;49(5):582\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eGaballa D, Bezinover D, Kadry Z, Eyster E, Wang M, Northup PG, et al.Development of a Model to Predict Portal Vein Thrombosis in Liver Transplant Candidates: The Portal Vein Thrombosis Risk Index. Liver Transplant. 2019;25(12):1747\u0026ndash;55.\u003c/li\u003e\n\u003cli\u003eStine JG, Wang J, Shah PM, Argo CK, Intagliata N, Uflacker A, et al.Decreased portal vein velocity is predictive of the development of portalvein thrombosis: A matched case-control study. Liver Int. 2018;38(1):94\u0026ndash;101.\u003c/li\u003e\n\u003cli\u003eFrancoz C, Belghiti J, Vilgrain V, Sommacale D, Paradis V, Condat B, etal. Splanchnic vein thrombosis in candidates for liver transplantation:usefulness of screening and anticoagulation. Gut. 2005;54(5):691\u0026ndash;7.\u003c/li\u003e\n\u003cli\u003eBasili S, Carnevale R, Nocella C, Bartimoccia S, Raparelli V, et al. Serum Albumin Is Inversely Associated With Portal Vein Thrombosis in Cirrhosis. Hepatol Commun. 2019;3(4):504\u0026ndash;12.\u003c/li\u003e\n\u003cli\u003eGiannitrapani L, Gran\u0026agrave; W, Licata A, Schiavone C, Montalto G, Soresi M. Nontumorous portal vein thrombosis in liver cirrhosis: Possible role of \u0026beta;-blockers. Med Princ Pract. 2018;27(5):466\u0026ndash;71.\u003c/li\u003e\n\u003cli\u003eYerdel M a, Gunson B, Mirza D, Karayal\u0026ccedil;in K, Olliff S, Buckels J, et al.Portal vein thrombosis in adults undergoing liver transplantation: riskfactors, screening, management, and outcome. Transplantation.2000;69(9):1873\u0026ndash;81.\u003c/li\u003e\n\u003cli\u003eXu X, Guo X, De Stefano V, Silva-Junior G, Goyal H, Bai Z, et al.Nonselective beta-blockers and development of portal vein thrombosis in liver cirrhosis: a systematic review and meta-analysis. Hepatology International.2019;13(4):468\u0026ndash;81.\u003c/li\u003e\n\u003cli\u003eLisman T, Porte RJ. Rebalanced hemostasis in patients with liverdisease: Evidence and clinical consequences. Blood. 2010;116(6):878\u0026ndash;85\u003c/li\u003e\n\u003cli\u003eKalambokis GN, Oikonomou A, Christou L, Baltayiannis G. High von Willebrand Factor Antigen Levels and Procoagulant Imbalance May BeInvolved in Both Increasing Severity of Cirrhosis and Portal VeinThrombosis. 2016;64(4):1383\u0026ndash;5.\u003c/li\u003e\n\u003cli\u003eLa Mura V, Tripodi A, Tosetti G, Cavallaro F, Chantarangkul V, Colombo M, et al. Resistance to thrombomodulin is associated with de novo portal vein thrombosis and low survival in patients with cirrhosis. Liver Int. 2016;36:1\u0026ndash;9.\u003c/li\u003e\n\u003cli\u003eMartinelli I, Primignani M, Aghemo A, Reati R, Bucciarelli P, Fabris F, etal. High levels of factor VIII and risk of extra-hepatic portal veinobstruction. J Hepatol. 2009;50(5):916\u0026ndash;22.\u003c/li\u003e\n\u003cli\u003eLancellotti S, Basso M, Veca V, Sacco M, Riccardi L, Pompili M, et al.Presence of portal vein thrombosis in liver cirrhosis is strongly associatedwith low levels of ADAMTS-13: a pilot study. Intern Emerg Med.2016;11(7):959\u0026ndash;67.\u003c/li\u003e\n\u003cli\u003eMa SD, Wang J, Bezinover D, Kadry Z, Northup PG, Stine JG. Inherited thrombophilia and portal vein thrombosis in cirrhosis: A systematic review and meta-analysis. Res Pract Thromb Haemost. 2019;3(4):658\u0026ndash;67.\u003c/li\u003e\n\u003cli\u003eMangia A, Villani MR, Cappucci G, Santoro R, Ricciardi R, Facciorusso D,et al. Causes of portal venous thrombosis in cirrhotic patients: The role ofgenetic and acquired factors. Eur J Gastroenterol Hepatol.2005;17(7):745\u0026ndash;51.\u003c/li\u003e\n\u003cli\u003eEngelmann B, Massberg S. Thrombosis as an intravascular effector ofinnate immunity. Nat Rev Immunol.2013;13(1):34\u0026ndash;45.\u003c/li\u003e\n\u003cli\u003eJim\u0026eacute;nez-Alc\u0026aacute;zar M, Kim N, Fuchs TA. Circulating Extracellular DNA:Cause or Consequence of Thrombosis Semin Thromb Hemost.2017;43(6):553\u0026ndash;61.\u003c/li\u003e\n\u003cli\u003eBerzigotti A, Piscaglia F. Ultrasound in Portal Hypertension \u0026ndash; Part 1.Ultraschall Med. 2011;32:548\u0026ndash;71.\u003c/li\u003e\n\u003cli\u003eLa Mura V, Reverter JC, Flores-Arroyo A, Raffa S, Reverter E, Seijo S, etal. Von Willebrand factor levels predict clinical outcome in patients with cirrhosis and portal hypertension. Gut. 2011;60(8):1133\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eRaffa S, Reverter JC, Seijo S, Tassies D, Abraldes JG, Bosch J, et al. Hypercoagulability in patients with chronic noncirrhotic portal vein thrombosis. Clin Gastroenterol Hepatol. 2012;10(1):72\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eAri\u0026euml;ns RAS, Kohler HP, Mansfield MW, Grant PJ. Subunit antigen and activity levels of blood coagulation factor XIII in healthy individuals:Relation to sex, age, smoking, and hypertension. Arterioscler ThrombVasc Biol. 1999;19(8):2012\u0026ndash;6.\u003c/li\u003e\n\u003cli\u003eHemker HC, Giesen P, Al Dieri R, Regnault V, De Smedt E, WagenvoordR, et al. Calibrated automated thrombin generation measurement in clotting plasma. Pathophysiol Haemost Thromb. 2003;33(1):4\u0026ndash;15.\u003c/li\u003e\n\u003cli\u003eMart\u0026iacute;nez-Zamora MA, T\u0026agrave;ssies D, Creus M, Reverter JC, Puerto B,Monteagudo J, et al. Higher levels of procoagulant microparticles in women with recurrent miscarriage are not associated with antiphospholipid antibodies. Hum Reprod. 2016;31(1):46\u0026ndash;52.\u003c/li\u003e\n\u003cli\u003eVon Meijenfeldt FA, Burlage LC, Bos S, Adelmeijer J, Porte RJ, Lisman T. Elevated Plasma Levels of Cell-Free DNA During Liver Transplantation Are Associated With Activation of Coagulation. Liver Transplant.2018;24(12):1716\u0026ndash;25.\u003c/li\u003e\n\u003cli\u003eLisman T, De Groot PG, Meijers JCM, Rosendaal FR. Reduced plasma fibrinolytic potential is a risk factor for venous thrombosis. Blood. 2005;105(3):1102\u0026ndash;5.\u003c/li\u003e\n\u003cli\u003eAleman MM, Byrnes JR, Wang JG, Tran R, Lam WA, Paola J Di, et al.Factor XIII activity mediates red blood cell retention in venous thrombi. J Clin Invest. 2014;124(8):3590\u0026ndash;600.\u003c/li\u003e\n\u003cli\u003eLisman T, Moschatsis S, Adelmeijer J, Karel Nieuwenhuis H, De Groot PG. Recombinant factor VIIa enhances deposition of platelets with congenital or acquired \u0026alpha;IIb\u0026beta;3 deficiency to endothelial cell matrix and collagen under conditions of flow via tissue factor-independent thrombin generation. Blood. 2003;101(5):1864\u0026ndash;70.\u003c/li\u003e\n\u003cli\u003eLisman T, Ari\u0026euml;ns RAS. Alterations in Fibrin Structure in Patients with Liver Diseases. Semin Thromb Hemost. 2016;42(4):389\u0026ndash;96.\u003c/li\u003e\n\u003cli\u003eHugenholtz GCG, Macrae F, Adelmeijer J, Dulfer S, Porte RJ, Lisman T,et al. Procoagulant changes in fibrin clot structure in patients withcirrhosis are associated with oxidative modifications of fibrinogen. J Thromb Haemost. 2016;14(5):1054\u0026ndash;66\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Machine learning-based model, Portal vein thrombosis, Risk, Hepatitis B-related cirrhosis","lastPublishedDoi":"10.21203/rs.3.rs-3831874/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3831874/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"We aimed to assess the risk of portal vein thrombosis (PVT) in patients with hepatitis B-related cirrhosis (HBC) using artificial neural networks (ANN). PVT can exacerbate portal hypertension and lead to complications, increasing the risk of mortality. Unfortunately, accurate prediction models for PVT in hepatitis B cirrhosis patients are currently insufficient. To address this gap, we conducted a study at Beijing Ditan Hospital, affiliated with Capital Medical University, involving 986 hospitalized patients. The patients were randomly divided into a training set (685 cases) and a validation set (301 cases) using a 3:1 ratio. Through univariate analysis, we determined independent factors that influence the occurrence of PVT, which were then utilized to develop an ANN model. The performance of the ANN model was assessed using various indicators, such as the area under the receiver operating characteristic curve (AUC) and concordance index (C-index). In the training group, PVT developed within three years in 19.0% of patients, and within five years in 23.7% of patients. Similarly, in the validation group, PVT developed within three years in 16.7% of patients, and within five years in 24.0% of patients. The ANN model incorporated nine independent risk factors, including age, presence of ascites, manifestation of hepatic encephalopathy (HE), occurrence of gastrointestinal varices with bleeding, Child-Pugh classification, alanine transaminase (ALT) levels, albumin (ALB) levels, neutrophil-to-lymphocyte ratio (NLR), and platelet count (PLT). Importantly, the AUC of the ANN model was significantly higher at 0.9718 compared to existing models such as MELD and CTP (all p\u003c0.001). Our ANN model effectively classified patients into high ,medium, and low risk groups for PVT development over a span of 3 and 5 years. These findings were further validated in an independent cohort.","manuscriptTitle":"Machine learning-based model used for predicting Portal vein thrombosis risk for patients with cirrhosis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-10 21:41:50","doi":"10.21203/rs.3.rs-3831874/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"003ffd39-0e04-4032-9a49-0ab72b384f2d","owner":[],"postedDate":"January 10th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-06-04T11:14:46+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-10 21:41:50","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3831874","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3831874","identity":"rs-3831874","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}