模糊聚类和logistic回归对试管婴儿成功因素的分析研究

发布时间：2018-06-13 04:35

本文选题：试管婴儿 + logistic回归模型　；参考：《中国卫生统计》2015年02期

【摘要】：目的分析影响试管婴儿成功率的关键因素,评价关键因素的相对重要性。方法通过采集试管婴儿医疗数据,确定影响因素并进行离散化,建立了妊娠结果对影响因素的logistic回归模型。运用优势分析法定量评价关键因素的相对重要性。最后对所有样本进行加权模糊聚类,分为普通和高危两组患者并比较两组医学特征的异同。结果对logistic回归模型进行统计学假设检验,得到年龄、移植胚胎数、促排卵药物(Gn)总量、输卵管和卵巢为影响成功率的关键因素。其重要性顺序依次为:移植胚胎数、年龄、促排卵药物(Gn)总量、输卵管和卵巢。对聚类数据进行统计学假设检验,得到两组患者在不孕状态、子宫和卵巢方面差异具有统计学意义,在盆腔、输卵管、子宫内膜异位和精子质量方面差异不具有统计学意义。结论 logistic回归模型适用于分析影响试管婴儿成功率的关键因素。
[Abstract]:Objective to analyze the key factors influencing the success rate of IVF and evaluate the relative importance of the key factors. Methods by collecting the medical data of IVF, the influencing factors were determined and discretized, and the logistic regression model of the influencing factors was established. The relative importance of key factors is evaluated quantitatively by using advantage analysis method. Finally, all the samples were grouped by weighted fuzzy clustering and divided into two groups: common and high risk groups, and the similarities and differences of medical characteristics between the two groups were compared. Results the logistic regression model was tested with statistical hypothesis. The key factors affecting the success rate were age, the number of embryos transferred, the total amount of ovulation promoting drug Gnn, and the oviduct and ovary. The order of importance is the number of embryos transferred, age, the total amount of ovulation stimulating drug Gnn, oviduct and ovary. The statistical hypothesis test of cluster data showed that there were significant differences in infertility, uterus and ovary between the two groups, but there was no significant difference in pelvic cavity, fallopian tube, endometriosis and sperm quality. Conclusion logistic regression model can be used to analyze the key factors influencing the success rate of IVF.
【作者单位】：宁波大红鹰学院;
【基金】：浙江省教育厅科研项目(Y201329710) 宁波市自然科学基金(2013A610115)
【分类号】：R714.8

【参考文献】