Lightweight Deep Learning Classification Model for Identifying Low-Resolution CT Images of Lung Cancer

Table 4

Performance comparison of different parameters of the two-dimensional DenseNet model.

Deep learning network	Accuracy	Recall	Precision	AUC
Two-dimensional DenseNet [5]	76.67	63.33	90	0.8889
Two-dimensional DenseNet-bottleneck layer [8]	54.17 (↓22.5)	20.84 (↓42.49)	87.50 (↓2.50)	0.5926 (↓0.2963)
Two-dimensional DenseNet-no waiver [10]	58.33 (↓18.34)	33.33 (↓30.00)	83.33 (↓6.67)	0.6833 (↓0.205 6)
Two-dimensional DenseNet-no data extension	58.33 (↓18.34)	63.33 (↓0.00)	53.34 (↓36.66)	0.6944 (↓0.1945)