论文部分内容阅读
目的:比较支持向量机(support vector machine,SVM)和传统的Logistic回归构建的急性出血性脑卒中(intracerebral hemorrhage,ICH)早期预后判别模型的预测性能,探索急性ICH预后研究的新方法。方法 :收集急性ICH患者339例,随访观察21 d时的临床转归情况。应用随机数字法以3∶1的比例分为两组,一组作为训练样本用于筛选变量和建立预测模型,计254例;另一组作为验证样本,用于评价模型预测效果,计85例。建模方法采用SVM和常规统计方法中的Logistic回归。结果:通过对85例ICH患者的预测判别验证,SVM1的预测分类能力在4个模型为最强,4个模型预测的准确率和Youden指数分别为:Logistic回归:72.9%(62.0%~81.7%)、0.441(0.249~0.633);SVM1:82.4%(72.3%~89.5%)、0.632(0.465~0.799);SVM2:78.8%(68.4%~86.6%)、0.557(0.379~0.735);SVM3:78.8%(68.4%~86.6%)、0.563(0.385~0.741)。结论:采用SVM能较好地判断急性ICH患者的早期预后,其效能优于Logistic回归模型。
OBJECTIVE: To compare the predictive performance of the early prognosis model of acute hemorrhagic stroke (ICH) with support vector machine (SVM) and traditional Logistic regression, and to explore a new method for the prognosis of acute ICH. Methods: A total of 339 patients with acute ICH were collected and followed up for 21 days. Random number method was used to divide the patients into three groups according to the ratio of 3:1. One group was used as the training sample to select the variables and establish the predictive model, with 254 cases in total. The other group was used as the verification sample to evaluate the predictive effect of the model. . The modeling method uses Logistic regression in SVM and conventional statistical methods. Results: The predictive classification of 85 ICH patients showed that SVM1 had the strongest prediction classification ability among the four models. The prediction accuracy of the four models and the Youden index were: Logistic regression: 72.9% (62.0% -81.7% ), 0.441 (0.249-0.633), SVM1: 82.4% (72.3% -89.5%), 0.632 (0.465-0.799), SVM2: 78.8% (68.4% -86.6%), 0.557 (0.379-0.735) % (68.4% ~ 86.6%), 0.563 (0.385 ~ 0.741). Conclusion: The SVM can be used to judge the early prognosis of patients with acute ICH better than Logistic regression model.