Bayesian Network to Predict Hepatitis B Surface Antigen Seroclearance in Chronic Hepatitis B Patients
Yun Huang 1 , Xiangyong Li 2 , Xiaoyan Zheng 2 , Xiaolu Tian 3 , Mingxue Yu 2 , Liuping Sha 2 , Yinhui Liu 4 , Yutian Chong 2 , Yuantao Hao 1 , Xu You 5
Affiliations
Affiliations
1
Department of Medical Statistics, School of Public Health, Sun Yat-Sen University, No. 74 Zhongshan II Road, Guangzhou, 510080, Guangdong Province, People's Republic of China.
2
Department of Infectious Diseases, the Third Affiliated Hospital, Sun Yat-Sen University, No. 600 Tianhe Road, Guangzhou, 510630, Guangdong Province, People's Republic of China.
3
Department of Big Data, China Construction Bank Fintech Co. Ltd.; Tower A, NEO building, No, 6009 Shennan Avenue, Shenzhen, 518048, Guangdong Province, People's Republic of China.
4
Department of Infectious Diseases, People's Hospital of Foshan Sanshui Dictrict, No.16 Southwest Guanghai Avenue West, Foshan, 528100, Guangdong Province, People's Republic of China.
5
Department of Clinical Laboratory, the Third Affiliated Hospital, Southern Medical University, No. 183 Zhongshan Avenue West, Guangzhou, 510630, Guangdong Province, People's Republic of China.
PMID: 32741055 DOI: 10.1111/jvh.13368
Abstract
There is a need for an interpretable, accurate and interactions-considered model for predicting hepatitis B surface antigen (HBsAg) seroclearance.We aimed to construct a Bayesian network (BN) model using available medical records to predict HBsAg seroclearance in chronic hepatitis B (CHB) patients, and to evaluate the model's performance.This was a case-control study. A total of 1,966 consecutive CHB patients (mean age 39.04 ± 11.23 years) between January 2006 and June 2015 were included. The demographic and clinical characteristics, laboratory data, and imaging parameters were obtained and used to build a BN model to estimate the probability of HBsAg seroclearance. Baseline serum HBsAg and hepatitis Be antigen (HBeAg) levels, virological response, and HBeAg seroclearance were the most significant predictors of HBsAg seroclearance. The post-test probability table showed that patients with baseline HBsAg concentrations ≤ 2,000 IU/mL, negative baseline HBeAg, an initial virological response, and without HBeAg seroclearance (i.e. no recurrence of HBeAg positivity during follow-up) were most likely to have HBsAg seroclearance. The constructed BN model had an area under the receiver operating characteristic curves of 0.896 (95% confidence interval [CI]: 0.892, 0.899), a sensitivity of 0.840 (95% CI: 0.833, 0.846), a specificity of 0.880 (95% CI: 0.876, 0.884), and an accuracy of 0.878 (95% CI: 0.874, 0.882) for predicting HBsAg seroclearance.The established BN model accurately estimated the probability of HBsAg seroclearance and is a promising tool to assist clinical decision making.
需要一种可解释，准确且相互作用考虑的模型来预测乙型肝炎表面抗原（HBsAg）血清清除率。我们旨在利用现有的医学记录构建贝叶斯网络（BN）模型，以预测慢性乙型肝炎（CHB）中的HBsAg血清清除率）患者，并评估模型的性能。这是一项病例对照研究。纳入2006年1月至2015年6月之间的1,966名连续性CHB患者（平均年龄39.04±11.23岁）。获得了人口统计学和临床特征，实验室数据和影像学参数，并将其用于建立BN模型以估计HBsAg血清清除的可能性。基线血清HBsAg和肝炎Be抗原（HBeAg）水平，病毒学应答和HBeAg血清清除率是HBsAg血清清除率的最重要预测指标。测试后概率表显示，基线HBsAg浓度≤2,000 IU / mL，基线HBeAg阴性，初次病毒学应答且无HBeAg血清清除（即随访期间HBeAg阳性无复发）的患者最有可能患有HBsAg清除血清。所构建的BN模型在接收器工作特性曲线下的面积为0.896（95％置信区间[CI]：0.892，0.899），灵敏度为0.840（95％CI：0.833，0.846），特异性为0.880（95％）。 CI：0.876，0.884），预测HBsAg血清清除率的准确度为0.878（95％CI：0.874，0.882），已建立的BN模型可以准确估计HBsAg血清清除的可能性，是有助于临床决策的有前途的工具。