Table 3.

Classification results of experiment 1: Using weighted class balancing technique.

Dataset	CNN	ACC (%)	SEN (%)	SPE (%)	PPV (%)(%)	NPV (%)	F1 (%)	AUC (%)
Dataset 1	DenseNet201 121	96.60	96.59	98.29	96.58	94.58	96.36	97.80
	CheXNet	98.87	98.86	99.43	100.00	98.61	98.21	99.15
	Xception	95.92	95.91	97.95	95.92	92.90	95.91	96.94
	ResNet152	97.51	97.50	98.75	97.51	95.22	97.50	98.13
	VGG19	98.41	98.41	99.20	98.41	96.43	98.41	98.81
	MobileNetV2	95.69	95.69	97.84	95.69	93.90	95.69	96.76
Dataset 2	DenseNet201	93.58	93.58	96.79	93.58	90.18	93.58	95.38
	CheXNet	91.82	91.82	95.91	97.27	95.70	73.00	94.09
	Xception	89.94	89.91	94.96	90.05	88.15	89.55	92.34
	ResNet152	87.55	87.56	93.77	89.29	90.29	90.86	93.81
	VGG19	92.97	92.87	96.48	92.92	91.86	92.97	91.21
	MobileNetV2	91.57	91.57	95.78	91.57	90.17	91.68	93.86
Dataset 3	DenseNet201	94.00	93.85	96.92	93.94	92.04	94.87	94.08
	CheXNet	92.97	92.96	96.48	92.96	90.41	92.93	93.06
	Xception	93.00	92.91	96.45	93.00	90.12	92.87	93.62
	ResNet152	90.43	90.43	95.21	90.42	89.37	90.37	89.00
	VGG19	94.87	94.87	97.43	94.92	92.98	94.87	94.86
	MobileNetV2	95.69	91.51	95.51	91.62	88.53	91.57	90.69