TABLE 4.

The accuracy, specificity, sensitivity, AUC, and AP score of different models at distinguishing Noonan syndrome from patients with several other genetic syndromes.

models	Accuracy (mean ± SD)	Specificity (mean ± SD)	Sensitivity (mean ± SD)	AUC (mean ± SD)	AP score (mean ± SD)
DCNN-Arcface	0.8171 ± 0.0074	0.9477 ± 0.0116	0.7794 ± 0.0252	0.9274 ± 0.0062	0.9356 ± 0.0067
DCNN-CE	0.7848 ± 0.0205	0.7907 ± 0.0155	0.6960 ± 0.0207	0.8594 ± 0.0106	0.8739 ± 0.0108
SVM-linear	0.7048 ± 0.0190	0.6982 ± 0.019	0.7112 ± 0.0049	0.7627 ± 0.0161	0.7499 ± 0.0257
LR	0.7210 ± 0.0111	0.7273 ± 0.0407	0.7150 ± 0.0294	0.7694 ± 0.0102	0.7467 ± 0.0025

Values in bold indicate the optimal performance. SD, standard deviation.