TABLE 3.

The latest artificial intelligence‐based methods for reducing the false positive rate.

Year	Authors	Method/identified features	Dataset	Quality index	Quality index value
2013	Tartar et al. 67	Shape features	Dataset from Cerrahpasa Medicine Faculty, Istanbul University	Sensitivity	0.896
				Specificity	0.875
2014	Teramoto et al. 107	Shape features, intensity	Cancer‐screening program at the East Nagoya Imaging Diagnosis Center	Sensitivity	0.83
2018	Gong et al. 29	Intensity, shape, texture features	LUNA16/ANODE09	Sensitivity	0.8462
2019	Zuo et al. 108	Multi‐resolution features integrated 2D CNN	LUNA16	Accuracy	0.9733
2019	Zhou et al. 95	2/3D Models Genesis with encoder‐decoder architecture	LUNA16	AUC	0.982
2019	Kim et al. 68	Multi‐scale gradual integration CNN	LUNA16	CPM	0.942
2020	Sun et al. 109	S‐transform	Dataset from Sichuan Provincial People's Hospital	Sensitivity	0.9787
2020	Zuo et al. 69	Multi‐branch 3D CNN	LUNA16	CPM	0.83
2020	Masood et al. 70	Multi‐PRN inspired by VGG‐16	LUNA16/LIDC‐IDRI	Sensitivity	0.974
2021	Majidpourkhoei et al. 110	CADe/CADx	LIDC‐IDRI	Accuracy	0.901
				Sensitivity	0.841
				Specificity	0.917
2021	Yuan et al. 71	MP‐3D‐CNN	LUNA16	CPM	0.881
				Sensitivity	0.962
2023	Mkindu et al. 111	3D residual CNN with 3D ECA	LUNA16	CPM	0.911
				Sensitivity	0.9865

Abbreviations: AUC, area under the curve; CNN, convolutional neural network; CPM, competitive performance metrics.