模糊判别分析在妊高征预测中的应用研究

发布时间：2018-05-08 01:25

本文选题：多元统计分析 + 模糊统计学　；参考：《东北师范大学》2007年硕士论文

【摘要】： 妊娠高血压综合征(简称妊高征)是妊娠期特有的疾病。发病率在我国为9.4%,国外为5%～12%,该病严重影响母婴健康,是孕产妇和围生儿患病及死亡的主要原因~([1])。妊高征的发病原因及病理变化学说颇多,但至今尚无定论。因此,目前妊高征尚无法对因治疗,临床处理主要采用早期诊断和对症治疗。由此可见,建立可靠的妊高征预测性诊断方法对降低其发病率,控制病情发展及减少对母婴的危害具有重要意义。目前国内外所进行的预测性诊断方法主要为单项指标(单因素)的预测,其敏感度、特异度较低或不稳定,因而其预测性诊断价值尚存在不足~([2])。妊高征的病因可能是多因素的综合作用~([3])。因此,用多因素多指标联合预测,可能大大提高预测方法的价值。目前应用多因素预测模型预测性诊断妊高征的研究仅见个别报道,而且模型涉及的因素少,也未对模型预测的效果作出评价。本文正是为了解决以上问题,提出了模糊判别分析模型,预测性诊断妊高征。本文首先介绍了妊高征的基本概念,研究背景及预测妊高征的重要意义,既而介绍了目前国内外预测妊高征的研究现状以及本文的研究内容。本文的中心部分是模糊数字特征的提出以及模糊判别分析模型的建立。为了证明模型的有效性,通过面向对象方法,设计并实现了FSAS系统。最后,通过大量实测数据,分别用多元统计判别分析模型和模糊判别分析模型对妊高征进行预测,将二者预测结果对比分析,证明模糊判别分析方法比经典的判别分析方法更准确,同时指出了未来可继续研究的内容。
[Abstract]:Pregnancy induced hypertension syndrome (PIH) is a special disease during pregnancy. The incidence rate is 9.4in our country and 50.12in foreign countries. The disease seriously affects maternal and child health and is the main cause of morbidity and mortality of pregnant women and perinatal children ([1]). There are many theories about the pathogenesis and pathological changes of PIH, but there is no conclusion yet. Therefore, at present, PIH can not be treated, clinical treatment is mainly used early diagnosis and symptomatic treatment. Therefore, it is important to establish a reliable predictive diagnostic method for PIH to reduce the incidence, control the development of PIH and reduce the harm to mother and child. At present, the predictive diagnostic methods at home and abroad are mainly single index (single factor) prediction, its sensitivity, specificity is low or unstable, so its predictive diagnostic value is still insufficient ([2]). The etiology of PIH may be caused by multiple factors ([3]). Therefore, the value of forecasting method may be greatly improved by using multiple factors and multiple indexes. At present, there are only a few reports on predictive diagnosis of PIH using multivariate prediction model, and the model has few factors involved, and the effect of model prediction is not evaluated. In order to solve the above problems, a fuzzy discriminant analysis model is proposed to predict the diagnosis of pregnancy induced hypertension (PIH). This paper first introduces the basic concept of PIH, the research background and the significance of PIH prediction, then introduces the current research situation and the research content of PIH at home and abroad. The central part of this paper is the presentation of fuzzy digital features and the establishment of fuzzy discriminant analysis model. In order to prove the validity of the model, the FSAS system is designed and implemented by object-oriented method. Finally, the multivariate statistical discriminant analysis model and fuzzy discriminant analysis model are used to predict PIH through a large number of measured data. It is proved that the fuzzy discriminant analysis method is more accurate than the classical discriminant analysis method.
【学位授予单位】：东北师范大学
【学位级别】：硕士
【学位授予年份】：2007
【分类号】：R714.2;R311

【参考文献】