不同中医证型复发脑梗死的眼底血管特征分析及预测模型的建立

发布时间：2018-02-03 07:39

本文关键词： 眼底血管复发脑梗死中医证型风险预测　出处：《辽宁中医杂志》2017年05期 　论文类型：期刊论文

【摘要】：目的:分析影响复发脑梗死的眼底血管特征参数,建立不同中医证型复发脑梗死的预测模型。方法:回顾性分析196例在本科住院的脑梗死患者的临床资料,其中初发脑梗死137例,复发脑梗死59例。采集所有患者视网膜图像后用自动化的视网膜图像分析系统进行眼底血管特征分析,建立不同中医证型脑梗死复发风险预警模型,评估该模型的预测价值,分析复发脑梗死的眼底血管影响因素。结果:利用logistic回归模型建模,对测试集样本预测,气虚血瘀型和风痰阻络型复发脑梗死模型预测正确率分别为86.3%和87.0%。结论:利用logistic回归分析建立预测模型可以预测不同中医证型脑梗死复发的风险;多种眼底血管特征与脑梗死复发相关。
[Abstract]:Objective: to analyze the characteristic parameters of ocular fundus vessels affecting recurrent cerebral infarction and to establish a predictive model of recurrent cerebral infarction with different TCM syndromes. Methods: the clinical data of 196 patients with cerebral infarction in undergraduate course were analyzed retrospectively. There were 137 cases of primary cerebral infarction and 59 cases of recurrent cerebral infarction. The retinal images of all the patients were collected and analyzed by automatic retinal image analysis system. To establish the early warning model of recurrence risk of cerebral infarction of different TCM syndromes, to evaluate the predictive value of the model, and to analyze the influencing factors of ocular fundus vessels of recurrent cerebral infarction. Results: the logistic regression model was used to model the model. Sample prediction for test set. The correct prediction rates of recurrent cerebral infarction models of qi deficiency and blood stasis type and wind-phlegm blocking collaterals type were 86.3% and 87.0, respectively. The risk of cerebral infarction recurrence in different TCM syndromes can be predicted by logistic regression analysis. Multiple fundus vascular features are associated with recurrence of cerebral infarction.
【作者单位】：深圳市中医院针灸科;香港中文大学公共卫生学院生物统计学;
【基金】：2014深圳市科技计划项目(JCYJ20140408152909288)
【分类号】：R277.7
【正文快照】： 我国大量流行病学研究证实了脑卒中一直是国内人群死亡的主要原因[1-3]。而在脑卒中分类构成比中,脑梗死的比例最高[4],其高发病率、高死亡率和高致残率给患者带来巨大的痛苦,给家庭和社会带来沉重的经济负担。近年来脑梗死的复发率呈上升趋势,复发性脑梗死的病死率、致残率均

【参考文献】