进展性脑梗死多因素回归分析及判别模型建立

摘要

Objective To establish a discriminant model of risk factors for progressive cerebral infarction, and to provide a relatively accurate scientific basis for its early diagnosis. Methods Through logistic regression analysis, key factors for progressive cerebral infarction were selected according to Bayes theory, as well as a discriminant model was established. Results Discriminant model was described as follow: Y1 = -30.381 ± 8.025X7 ± 1.341X8 ± 2.301X21 ± 3.931X31 ± 5.802X32 ± 0.933X33 ± 2.H9X34; Y2= -21.875 ± 6.834X7 ± 2.160X8 ± 3.328X31 ± 3.726X32 ± 2.016X21 ± 0.861X33 ± 1.807X34. Conclusion The discriminant model might be used to contribute to the early diagnosis, early intervention, and timely treatment on progressive cerebral infarction.%目的:建立进展性脑梗死判别预测模型,为早期发现和干预治疗提供科学依据.方法:使用Logistic逐步回归分析关键影响因素,采用逐步Bayes判别法对进展性脑梗死患者的关键因素进行筛选,并建立判别预测模型.结果:判别模型为:(1)Y1(进展)＝-30.381 ± 8.025 × X7 ±341 × X8 ±2.301 × X21 ±3.931 × X31±5.802 × X32 ± 0.933 × X33±2.119 × X34.(2)Y2(非进展)=-21.875±6.834 × X7±2.160 × X8±3.328 × X31±3.726 × X32 ± 2.016 × X21±0.861 × X33±1.807 × X34(X7为平均收缩压,X8为体温,X21为血糖,X31为FIB,X32为颈动脉斑块性质,X33为颈动脉狭窄程度,X34为梗死部位).经评价预测效果较好.结论:判别模型的建立有助于进展性脑梗死的早期诊断分析及早期干预和治疗.