Table 1.

Performance of the studies exploring detection of pulmonary nodules.

Detection
Author	Year	Deep Learning Architecture	Dataset for Training	Dataset for Testing	Sensitivity	Specificity	AUC	Accuracy
Suzuki, Kenji * [19]	2009	MTANN	Independent dataset A	Independent dataset B	97	N/A	N/A	N/A
Tajbakhsh, Nima et al. [20]	2017	CNN	Independent dataset	Independent dataset	100	N/A	N/A	N/A
		MTANN	Independent dataset	Independent dataset	100	N/A	N/A	N/A
Masood, Anum et al. [21]	2018	FCNN	LIDC-IDRI, RIDER, LungCT-diagnosis, LUNA16, LISS, SPIE challenge dataset and independent dataset	RIDER	74.6	86.5	N/A	80.6
				SPIE challenge dataset	81.2	83	N/A	84.9
				LungCT-diagnosis	82.5	93.6	N/A	89.5
				Independent dataset	83.7	96.2	N/A	86.3
Chen, Sihang et al. [22]	2019	CNN	Independent dataset	Independent dataset	97	N/A	N/A	N/A
Liao, Fangzhou et al. [23]	2019	CNN	LUNA16 and DSB17	DSB17	85.6	N/A	N/A	N/A
Liu, Mingzhe et al. [24]	2018	CNN	LUNA16 and DSB17	DSB17	85.6	N/A	N/A	N/A
Li, Li et al. * [17]	2018	CNN	LIDC-IDRI and NLST	Independent dataset	86.2	N/A	N/A	N/A
Wang, Yang et al. [25]	2019	RCNN	Independent dataset	Independent dataset	N/A	N/A	N/A	N/A
Setio, A.A.A et al. * [18]	2016	CNN	LIDC-IDRI and ANODE09	DLCST	76.5	N/A	N/A	94
				ANODE09	N/A	N/A	N/A	N/A
Wang, Jun et al. [26]	2019	CNN	Tianchi AI challenge dataset and independent dataset	Independent dataset	75.6	N/A	N/A	N/A

Studies marked with * are studies where test dataset was different from training dataset. AUC: area under the curve. Abbreviations: massive training artificial neural network (MTANN), convolutional neural network (CNN), lung image database consortium and image database resource initiative (LIDC-IDRI), reference image database to evaluate therapy response (RIDER), Society of Photo-Optical Instrumentation Engineers (SPIE), lung nodule analysis 2016 (LUNA16), lung CT imaging signs (LISS), Kaggle data science bowl 2017 (DSB17), Danish lung cancer screening trial (DLCST), automatic nodule detection 2009 (ANODE09).