Skip to main content

Table 6 Random forest models for predicting CIN occurrence

From: Establishment of multifactor predictive models for the occurrence and progression of cervical intraepithelial neoplasia

NO. Indicators Factors accuracy AUC OOB
8 All clinical features age + menopause+HPV + gravidity+parity+TCT 60.00 75.20 28.89
9 Significant genes TGFBR2+ FOXO1 + CTBP2 71.11 70.54 26.67
10 Significant genes + significant clinical features TGFBR2 + CTBP2 + FOXO1 + HPV + TCT 77.78 92.09 26.67
11 Genes as the risk factors in unvariable logistic analysis CTBP2 + FOXO1 73.33 74.51 40.00
12 Genes as the risk factors in unvariable logistic analysis + Significant genes CTBP2 + FOXO1 + HPV + TCT 84.44 90.51 22.22
13 Genes as the independent factors in multivariable logistic analysis CTBP2 / / /
14 Genes as the independent factors in multivariable logistic analysis CTBP2 + HPV + TCT 75.56 85.38 26.67