摘要
目的探讨Logistic回归模型和决策树模型对大面积急性岛叶梗死预后不良的预测价值。方法回顾性分析2019年1月—2022年12月收治的大面积急性岛叶梗死100例的临床资料,根据发病3个月后改良Rankin量表分为预后良好组和预后不良组。统计2组临床资料、实验室指标、疾病史等基线资料。构建Logistic回归模型、决策树模型分析预后不良影响因素。绘制受试者工作特征(ROC)曲线分析2种模型对预后不良的预测效能。结果发病3个月后失访3例,预后不良发生率为39.18%(38/97),预后良好率为60.82%(59/97)。Logistic回归分析显示,年龄、房颤、基线肌钙蛋白T(cTnT)、鸢尾素、闭锁蛋白(Occludin)是预后不良影响因素(P<0.01);决策树模型分析显示,年龄、房颤、基线cTnT及Occludin是预后不良影响因素;决策树模型预测预后不良的曲线下面积大于Logistic回归模型。结论年龄、房颤、基线cTnT、Occludin是大面积急性岛叶梗死患者预后不良影响因素,基于上述因素构建Logistic回归模型和决策树模型均具有良好应用价值,应结合2种模型优点,为临床诊治提供新思路。
Objective To explore the predictive value of Logistic regression model and decision tree model for poor prognosis of large area acute insular infarction.Methods Clinical data of 100 patients with large area acute insular infarction admitted from January 2019 to December 2022 were retrospectively analyzed.According to the modified Rankin scale at 3 months after the onset,they were divided into good prognosis group and poor prognosis group.Baseline data such as clinical data,laboratory indicators and disease history were collected.Logistic regression model and decision tree model were constructed to analyze the influencing factors of poor prognosis.Receiver operating characteristic(ROC) curve was drawn to analyze the predictive efficacy of the two models for poor prognosis.Results Three patients were lost to follow-up at 3 months after disease onset,and the incidence of poor prognosis was 39.18%(38/97),and good prognosis was 60.82%(59/97).Logistic regression analysis showed that age,atrial fibrillation,baseline troponin T(cTnT),irisin and Occludin were influencing factors of poor prognosis(P<0.01).Decision tree model analysis showed that age,atrial fibrillation,baseline cTnT and Occludin were influencing factors of poor prognosis.The are under the ROC curve(AUC) of decision tree model in predicting poor prognosis was greater than that of Logistic regression model.Conclusion Age,atrial fibrillation,baseline cTnT,and Occludin are poor prognostic factors for patients with large area acute insular infarction,and the Logistic regression model and decision tree model constructed based on the above factors have good application value.Therefore,the advantages of the two models should be combined to provide a new idea for clinical diagnosis and treatment.
作者
包曼
梁菲菲
孙子慧
BAO Man;LIANG Feifei;SUN Zihui(Department of General Medicine,the First Hospital of Zhangjiakou City,Zhangjiakou,Hebei 075000,China;Department of Geriatrics,the First Hospital of Zhangjiakou City,Zhangjiakou,Hebei 075000,China)
出处
《临床误诊误治》
CAS
2024年第3期45-51,共7页
Clinical Misdiagnosis & Mistherapy
基金
河北省卫生健康委办公室医学科学研究课题计划项目(202111605)
张家口市科学技术局市级科技计划项目(2121136D)。