Using Deep Learning for Classification of Lung Nodules on Computed Tomography Images

Table 5

Comparison with other papers.

Work	Database (samples)	Accuracy (%)	Sensitivity (%)	Specificity (%)
Nascimento et al. [21]	LIDC (73)	92.78	85.64	97.89
Orozco and Villegas [22]	NBIA-ELCAP (113)	N/A	96.15	52.17
Krewer et al. [7]	LIDC-IDRI (33)	90.91	85.71	94.74
Dandil et al. [23]	Private (128)	90.63	92.30	89.47
Parveen and Kavitha [24]	Private (3278)	N/A	91.38	89.56
Kuruvilla and Gunavathi, 2014 [6]	LIDC (110)	93.30	91.40	100
Gupta and Tiwari [25]	Private (120)	90	86.66	93.33
Hua et al. [10]	LIDC (2545)	N/A	73.30	78.70
Kumar et al. [8]	LIDC (4323)	75.01	83.35	N/A
da Silva [26]	LIDC-IDRI (8296)	82.3	79.4	83.8
CNN (this paper)	LIDC-IDRI (5024)	84.15%	83.96%	84.32%
DNN (this paper)	LIDC-IDRI (5024)	82.37%	80.66%	83.9%
SAE (this paper)	LIDC-IDRI (5024)	82.59%	83.96%	81.35%