Table 5.

Diagnostic performance of deep convolutional neural network model for differentiating triple-negative and non-triple-negative subtypes based on patient age in the main cohort.

Experiment	Model	AUC (%)	ACC (%)	SEN (%)	SPC (%)	YI (%)
Age ≤ 50	EfficientNet-B2	71.2	58.5	81.8	53.7	35.5
	DenseNet-121	68.5	63.1	81.8	59.3	41.1
	VGGNet-19	63.5	63.1	72.7	61.1	33.8
Age > 50	EfficientNet-B2	50.4	71.9	36.4	79.8	16.2
	DenseNet-121	51.2	33.1	95.5	19.2	14.6
	VGGNet-19	55.1	68.6	40.9	74.7	15.7

AUC, area under the curve; ACC, accuracy; SEN, sensitivity; SPC, specificity; YI, Youden’s index.