3D CNN with Visual Insights for Early Detection of Lung Cancer Using Gradient-Weighted Class Activation

Table 4

Key matrices with definitions.

Total population		True condition
		Condition positive	Condition negative
Predicted condition	Predicted condition positive	True positive	False positive (type I error)
	Predicted condition negative	False negative (type II error)	True negative
		True positive rate (TPR), recall, Sensitivity = ∑true positive/∑condition positive	False-positive rate (FPR) = ∑false positive/∑condition negative
		Accuracy (ACC) = (∑true positive + ∑true negative)/∑total population	Precision = ∑true positive/∑predicted condition positive
		F1 score = 1/((1/recall  + 1/precision)/2)	Specificity (SPC), selectivity = ∑true negative/∑condition negative