Table 2.

COVID-19/non-COVID-19 classification results. Class.: classification; Sens.: sensitivity; Spec.: specificity; Prec.: precision; Acc.: accuracy; AUC: area under the curve; Ref.: reference.

Class.	Subjects	Dataset	Method	Sens. (%) or recall	Spec. (%)	Prec. (%)	Acc. (%)	AUC (%)	F1-score	Ref.
COVID-19/ non-COVID-19	496 COVID-19 1385 others	Private + [28, 31]	CNN	94.06	95.47	N/A	94.98	97.91	NA	Jin et al. [30] Preprint
COVID-19/ non-COVID-19	N/A	[46]	CNN	~90	~90	N/A	~90	Not clear	~90	Singh et al. [32] Peer-reviewed
COVID-19/ non-COVID-19	449 COVID-19 100 normal 98 lung cancer 397 other	Private + [47, 48]	DL multitask	94	79	N/A	86	93	N/A	Amyar et al. [33] Preprint
COVID-19/ non-COVID-19	349 COVID-19 397 non-COVID-19	Private + [47, 49]	ResNet18	100.0	98.6	N/A	99.4	99.65	99.5	Ahuja et al. [34] Peer-reviewed
COVID-19/ non-COVID-19	564 COVID-19 660 non-COVID-19	[50]	VGG16 based lesion-attention DNN	88.8	N/A	87.9	88.6	94.0	87.9	Liu et al. [35] Conference proceeding
COVID-19/ non-COVID-19	53 COVID-19 97 other	Not clear	SVM	97.56	99.68	99.62	98.71	N/A	98.58	Barstugan et al. [36] Preprint
COVID-19/ non-COVID-19	313 COVID-19 229 without COVID-19	Private	UNet	90.7	91.1	N/A	90.1	95.9	N/A	Wang et al. [37] Peer-reviewed
COVID-19/ non-COVID-19	51 COVID-19 55 control	Private	UNet++	94.34	99.16	N/A	98.85	N/A	N/A	Chen et al. [38] Preprint
COVID-19/ non-COVID-19	723 COVID-19 413 others	Private	UNet++ + ResNet-50	97.4	92.2	N/A	N/A	99.1	N/A	Jin et al. [39] Preprint
COVID-19/ non-COVID-19	413 COVID-19 439 non-COVID-19	[32, 51]	ResNet-50 + 2D CNN	91.46	94.78	95.19	93.02	N/A	N/A	Pathak et al. [42] Peer-reviewed
COVID-19/ non-COVID-19	460 COVID-19 397 non-COVID-19	[26, 47]	CNN SqueezeNet	85.00	81.00	81.73	83.00	N/A	83.33	Polsinelli et al. [43] Preprint
COVID-19/ non-COVID	230 COVID-19 130 normal	Private	AD3D-MIL	97.9	NA	97.9	97.9	99.0	97.9	Han et al. [44] Peer-reviewed
COVID-19/ non-COVID	1029 COVID-19 1695 non-COVID-19	Private	AH-Net DenseNet121	84.0	93.0	NA	90.8	94.9	NA	Harmon et al. [45] Peer-reviewed