Skip to main content

Table 6 Characteristics of both combined models for diagnosis of EOC

From: A combined blood based gene expression and plasma protein abundance signature for diagnosis of epithelial ovarian cancer - a study of the OVCAD consortium

Model

L1 lasso penalty

L2 ridge penalty

Blood expression values

5 genes

13 genes

Plasma protein values

5 proteins

6 proteins

AUCa (FIGO I-IV)

0.998

0.998

AUC (FIGO I-II)

0.976

0.979

Specificity (set)

99.6%

99.6%

Sensitivity

97.8%

95.6%

Classification error (bootstrap .632+)

3.1%

2.8%

  1. aBootstrap .632+ validated area under the ROC curve.