摘要
目的通过建立随机森林模型预测老年患者大脑中动脉(middle cerebral artery,MCA)瘤的破裂风险。方法选取温州医科大学附属第一医院2009年3月至2020年6月收治的MCA动脉瘤老年患者(年龄>60岁)资料进行回顾性分析,按7∶3随机分为训练组和验证组,采用多因素Logistic回归分析老年患者MCA动脉瘤破裂的独立危险因素,据此建立随机森林模型预测动脉瘤破裂风险,并通过其他4家医院的数据进行外部验证。采用受试者操作特征曲线下面积(area under the curve,AUC)评估模型的预测效能。结果共纳入242例MCA动脉瘤(训练组169例,内部验证组73例)和外部验证组48例。多因素Logistic回归分析显示,尺寸比、动脉瘤角度、高宽比及不规则形态是老年患者MCA动脉瘤破裂的独立危险因素。随机森林模型对训练组、内部验证组和外部验证组预测效能的AUC分别为0.916(95%CI:0.878~0.946)、0.925(95%CI:0.874~0.968)和0.834(95%CI:0.725~0.932)。结论预测老年患者MCA动脉瘤破裂的随机森林模型具有较好的性能,可辅助临床诊疗决策。
Objective To predict the risk of middle cerebral artery(MCA)aneurysm rupture in elderly patients by establishing a random forest model.Methods A retrospective analysis was conducted on data from elderly patients(age>60)with MCA aneurysms treated at the First Affiliated Hospital of Wenzhou Medical University from March 2009 to June 2020.The data were randomly divided into a training group and a validation group at a 7∶3 ratio.Independent risk factors for MCA aneurysms rupture in elderly patients were obtained by unifactorial and multifactorial Logistic regression,based on these a random forest model was constructed,which was externally validated using data from four other hospitals.Its predictive performance was evaluated using the area under the curve(AUC)of receiver operating characteristic.Results A total of 242 MCA aneurysms were included(with 169 cases in training group,73 cases in internal validation group),and 48 cases in the external validation group.Multifactorial Logistic regression analysis showed that the size ratio,aneurysm angle,height-width ratio,and irregular morphology were independent risk factors for MCA aneurysms rupture in elderly patients.The random forest model achieved AUC values of 0.916(95%CI:0.878–0.946),0.925(95%CI:0.874–0.968),and 0.834(95%CI:0.725–0.932)for training,internal validation,and external validation groups,respectively.Conclusion The random forest model demonstrated excellent performance in predicting the risk of MCA aneurysm rupture in elderly patients and can be used to assist clinical decision-making.
作者
王殊
陈勇春
郑葵葵
周甲丰
WANG Shu;CHEN Yongchun;ZHENG Kuikui;ZHOU Jiafeng(Department of Radiology,The First Affiliated Hospital of Wenzhou Medical University,Wenzhou 325015,Zhejiang,China)
出处
《中国现代医生》
2025年第36期9-14,共6页
China Modern Doctor
基金
温州市科技计划项目(Y20220072)。
关键词
颅内动脉瘤
老年
大脑中动脉
破裂
随机森林
Intracranial aneurysm
Elderly
Middle cerebral artery
Rupture
Random forest
