Table 3.

COVID-19/non-COVID-19 pneumonia classification results. Class.: classification; bac. pneu.: bacterial pneumonia; 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/influ-A/normal	219 COVID-19 224 influ-A 175 normal	Private	CNN ResNet	86.7	N/A	81.3	N/A	N/A	83.9	Xu et al. [52] Peer-reviewed
COVID-19/CT-EGFR	1266 COVID-19 4106 CT-EGFR	Private	COVID19Net (DenseNet-like str.)	79.35	71.43	N/A	85.00	86.00	90.11	Wang et al. [53] Peer-reviewed
COVID-19/other pneu.	521 COVID-19 397 normal 76 bac. pneu. 48 SARS	[26, 47]	DL ShuffleNet V2	85.71	84.88	N/A	85.40	92.22	N/A	Hu et al. [10] Preprint
COVID-19/other pneu.	521 COVID-19 665 non-COVID-19 pneu.	Private	DNN EfficientNet B4	95	96	N/A	96	95	N/A	Bai et al. [54] Peer-reviewed
COVID-19/CAP	1495 COVID-19 1027 CAP	Private	Multiview representation learning	96.6	93.2	N/A	95.5	NA	N/A	Kang et al. [56] Peer-reviewed
COVID-19/CAP	1658 COVID-19 1027 CAP	Private	RF-based ML model	90.7	83.3	N/A	87.9	94.2	N/A	Shi et al. [57] Preprint
COVID-19/bac. pneu./normal	88 COVID-19 101 bac. pneu. 86 normal	Private	DRE-Net	96	N/A	79	86	95	87	Ying et al. [58] Preprint
COVID-19/other pneu./non-pneu.	230 COVID-19 100 normal	Private	AD3D-MIL	90.5	NA	95.9	94.3	98.8	92.3	Han et al. [44] Peer-reviewed
COVID-19/other pneu./non-pneu.	1194 COVID-19 1357 other pneu. 998 normal 444 lung cancer	Private + [26, 47]	FCONet ResNet50	99.58	100.0	NA	99.87	100.0	NA	Ko et al. [59] Peer-reviewed
COVID-19/other pneu./non-pneu.	1292 COVID-19 1735 pneumonia 713 non-pneu.	Private	COVNet ResNet50	90	96	NA	NA	96.0	NA	Li et al. [8] Peer-reviewed
COVID-19/other pneu./healthy	3854 COVID-19 6871 other pneu. 8566 healthy	Private	MVPNet 3D UNet 3D UNet-based network	100	25	NA	94	NA	97.0	Ni et al. [60] Peer-reviewed