Molecular differential diagnosis of follicular thyroid carcinoma and adenoma based on gene expression profiling by using formalin-fixed paraffin-embedded tissues

Table 2 Performance measures of classifiers in different datasets

Dataset (origin, method of analysis)	Method	Accuracy (%)	PPV (%)	NPV (%)	Sensitivity (%)	Specificity (%)	Prevalence of FTC in dataset (%)
B (own, microarray)	DLDA classification based on the 8 best genes chosen from 99 preselected ones.*	80	82	78	76	83	50
B (own, microarray)	DLDA classification based on 45 (optimal number) best genes chosen from 99 preselected ones.*	84	85	83	83	85	50
C (own, FFPE qPCR)	5-gene DLDA classification (cut-off 0.5)**	72(95% CI: 60–82)	67(95% CI: 48–82)	76(95% CI: 60–89)	71(95% CI: 52–86)	72(95% CI: 56–85)	44
C (own, FFPE qPCR)	5-gene DLDA classification (cut-off 0.12)**	70	61	88	90	55	44
D (own, microarray)	5-gene DLDA classifier trained on dataset B, tested on D	73	77	69	71	75	54
E1 (Weber et al. microarray)	5-gene DLDA classifier.**	92	100	86	83	100	50
E2 (Hinsch et al. microarray)	5-gene DLDA classifier.**	83	100	67	75	100	67

Accuracy, proportion of all samples that are correctly classified; PPV, positive predictive value; NPV, negative predictive value; SVM, support vector machines; DLDA, diagonal linear discriminant analysis; CI, confidence interval. *Performance assessed by 10-fold cross-validation. **Performance assessed by leave-one-out cross-validation.

ISSN: 1755-8794