预测输尿管软镜碎石术后并发尿源性脓毒症的反向传播神经网络模型构建  

Construction of a back propagation neural network model for predicting urosepsis after flexible ureteroscopic lithotripsy

作　　者：陈文炜 何彦丰[1,2] 卢凯鑫 刘昌毅 江涛[1,2] 张华[1,2] 高锐[1,2] 薛学义[1,2] CHEN Wenwei;HE Yanfeng;LU Kaixin;LIU Changyi;JIANG Tao;ZHANG Hua;GAO Rui;XUE Xueyi(Department of Urology,the First Affiliated Hospital of Fujian Medical University,Fuzhou 350005,China;Department of Urology,Binhai Campus of the First Affiliated Hospital of Fujian Medical University,National Regional Medical Center,Fuzhou 350212,China)

机构地区：[1]福建医科大学附属第一医院泌尿外科,福建福州350005 [2]福建医科大学附属第一医院滨海院区泌尿外科国家区域医疗中心,福建福州350212

出　　处：《浙江大学学报(医学版)》2025年第1期99-107,I0032-I0034,共12页Journal of Zhejiang University(Medical Sciences)

摘　　要：目的:构建输尿管软镜碎石术(FURL)后并发尿源性脓毒症的反向传播神经网络预测模型。方法:纳入428例接受FURL的肾结石患者,根据术后是否并发尿源性脓毒症分为脓毒症组(42例)和对照组(386例)。采用logistic回归分析确定FURL后并发尿源性脓毒症的影响因素及其交互作用。同时建立logistic回归模型和神经网络模型进行预测,通过受试者工作特征曲线评估两种模型的预测效能。结果:单因素分析显示,结石手术史、性别、尿培养阳性、结石直径、糖尿病、手术时间、白细胞、血小板、C反应蛋白(CRP)及肝素结合蛋白(HBP)水平与FURL后并发尿源性脓毒症显著相关(均P<0.05)。多因素分析表明,尿培养阳性、CRP及HBP水平是FURL后并发尿源性脓毒症的独立危险因素(均P<0.05)。交互作用分析显示,CRP与HBP对FURL后并发尿源性脓毒症的影响在相加模型(RERI=8.453,95%CI:2.645~16.282;AP=0.696,95%CI:0.131~1.273;S=3.369,95%CI:1.176~7.632)和相乘模型(OR=1.754,95%CI:1.218~3.650)中存在交互作用;CRP与尿培养对FURL后并发尿源性脓毒症的影响在相乘模型(OR=2.449,95%CI:1.525~3.825)中存在交互作用。预测模型比较显示,反向传播神经网络模型较logistic回归模型具有更优的预测效能。结论:CRP和HBP水平是FURL后并发尿源性脓毒症的独立危险因素,基于CRP、HBP等因素构建的反向传播神经网络模型较logistic回归模型具有更高的预测准确性。Objective:To analyze the association of serum heparin-binding protein(HBP)and C-reactive protein(CRP)levels with urosepsis following flexible ureteroscopic lithotripsy(FURL)and to construct a back propagation neural network prediction model.Methods:A total of 428 patients with kidney stones who underwent FURL were enrolled.Patients were divided into sepsis group(n=42)and control group(n=386)according to whether post-operative urosepsis developed.Logistic regression analysis was used to determine the risk factors of post-FURL urosepsis and their interactions.A logistic regression model and a back propagation neural network model were developed for predicting post-FURL urosepsis following FURL,and their predictive performance was evaluated using receiver operating characteristic curves.Results:Univariate analysis showed that stone surgery history,gender,positive urine culture,stone diameter,diabetes,operation time,white blood cell(WBC),platelet,CRP,and HBP levels were significantly associated with post-FURL urosepsis(all P<0.05).Multivariate analysis identified positive urine culture,CRP,and HBP levels as independent risk factors for post-FURL urosepsis(all P<0.05).Interaction analysis revealed that CRP and HBP showed both additive(RERI=8.453,95%CI:2.645-16.282;AP=0.696,95%CI:0.131-1.273;S=3.369,95%CI:1.176-7.632)and multiplicative(OR=1.754,95%CI:1.218-3.650)interactions,while CRP and urine culture demonstrated multiplicative interaction(OR=2.449,95%CI:1.525-3.825).The back propagation neural network model demonstrated superior predictive performance compared to the logistic regression model.Conclusions:CRP and HBP levels are independent risk factors for post-FURL urosepsis.The back propagation neural network model based on CRP and HBP exhibits higher predictive accuracy than the logistic regression model,which may provide a reliable risk assessment tool for early discrimination and intervention of post-FURL urosepsis.

关 键 词：肝素结合蛋白 C反应蛋白 输尿管软镜碎石术 尿源性脓毒症 预测 LOGISTIC回归模型 反向传播神经网络模型 

分 类 号：R691.3[医药卫生—泌尿科学]