基于基因表达子宫内膜癌的分类

发布时间：2018-05-12 01:02

本文选题：子宫内膜癌 + 基因表达　；参考：《数学的实践与认识》2017年18期

【摘要】：建立了四类基于基因表达的分类器,用以将87名妇女的子宫内膜样本分成癌症患者和非癌症患者.首先利用信噪比过滤掉无关基因,然后利用主成分分析降低样本维数,再针对这四类分类器随机取75个样本作为训练样本,其余的12个样本作为测试样本,实验结果表明这四类分类器适合子宫内膜癌的分类.最后采用留一交叉验证作为评判标准,通过比较,说明5BP-ELMAN分类器是一类更适合子宫内膜癌分类的有效的肿瘤分类器.
[Abstract]:Four types of classifier based on gene expression were established to divide the endometrium samples of 87 women into cancer patients and non cancer patients. First, the unrelated genes were filtered out by the signal to noise ratio. Then the sample dimension was reduced by principal component analysis, and 75 samples were randomly selected as training samples for the four classifiers, and the remaining 12 samples were taken as the training samples. As a test sample, the experimental results show that these four classifiers are suitable for the classification of endometrial cancer. Finally, the remaining one cross validation is used as a criterion. By comparison, the 5BP-ELMAN classifier is an effective tumor classifier which is more suitable for the classification of endometrial cancer.

【作者单位】：中北大学理学院;
【基金】：国家自然科学基金(61275120)
【分类号】：R737.33

【相似文献】