基于常规实验室检验结果的布鲁菌病预测模型的建立  

Establishment of Prediction Model of Brucellosis Based on Routine Laboratory Test Results

作　　者：杨芳[1] 王亚辉[1] 程林[1] 岳奇 杨萍[1] 何玲玲 Yang Fang;Wang Yahui;Cheng Lin;Yue Qi;Yang Ping;He Lingling(Laboratory of the General Hospital of the third division of Xinjiang production and Construction Corps,Tumushuke Xinjiang 843901,China)

机构地区：[1]新疆生产建设兵团第三师总医院检验科,新疆图木舒克市843901

出　　处：《医疗装备》2023年第1期9-12,18,共5页Medical Equipment

摘　　要：目的 利用常规临床数据建立模型对布鲁菌病进行预测诊断。方法 回顾性分析2019年4月至2022年2月医院收治的321例布鲁菌病患者为布鲁菌病组,选取同期261名健康人员为对照组,收集两组的人口学特征及实验室检查数据作为候选变量,对布鲁菌病的预测因素进行Logistic回归分析,确定独立预测因素,通过逐步回归法建立布鲁菌病预测模型和列线图,采用Bootstraping法对模型进行内部验证,并进行临床决策曲线分析。结果 Logistics分析结果表明,性别、C-反应蛋白(CRP)、谷丙转氨酶(ALT)、血清白蛋白(ALB)、血红蛋白(Hb)、血小板(PLT)水平为布鲁菌病的独立预测因素。逐步回归模型的曲线下面积(AUC)显著高于单变量CRP;校准曲线提示预测概率与实际概率具有较好的一致性;内部验证模型的AUC为0.93。结论 该研究利用布鲁菌病患者的临床特征和常规实验室检查数据建立了布鲁菌病的预测诊断模型,为临床医师评估布鲁菌病风险和预测诊断提供了重要参考依据。Objective By using routine clinical data, the model was established to predict the diagnosis of brucellosis. Methods With 321 patients with brucellosis admitted to the hospital from April 2019 to February 2022 as the brucellosis group, and 261 healthy people in the same period as the control group, by retrospective analysis, demographic characteristics and laboratory examination data of the two groups were collected as candidate variables.The predictive factors of brucellosis were analyzed by logistic regression, and the independent predictive factors were determined. The brucellosis prediction model and nomogram were established by stepwise regression method, the model was internally validated by Bootstrapping method, and the clinical decision curve analysis was conducted. Results Logistic analysis results showed that gender, C-reactive protein(CRP), alanine aminotransferase(ALT), serum albumin(ALB), hemoglobin(Hb) and platelet(PLT) were independent predictors of brucellosis.The area under curve(AUC) of stepwise regression model was significantly higher than that of univariate CRP;The predicted probability obtained from the calibration curve was in good agreement with the actual probability;AUC of internal validation model was 0.93.Conclusion In the study, the predictive diagnosis model of brucellosis was established,by using the clinical characteristics of patients with brucellosis and routine laboratory test data,which provided an important reference for clinicians to assess the risk of brucellosis and predict the diagnosis.

关 键 词：布鲁菌病 预测模型 诊断 常规实验室检查 

分 类 号：R155.3[医药卫生—营养与食品卫生学]