Deep-Learning-Based CT Imaging in the Quantitative Evaluation of Chronic Kidney Diseases

Table 1

Average value of the evaluation index of different deep network models on the test set (%).

Model		FCN-8S^[24]	FCN-16S^[24]	FCN-VGG10^[19]	U-Net^[27]	U-Net (+BN)	UNet++^[32]	RDA-UNET
Left kidney	DSC	88.12	87.53	95.23	92.65	95.09	94.88	96.25
	Precision	89.25	87.22	96.45	92.48	95.67	95.22	96.34
	Recall	89.88	87.43	95.55	93.66	96.45	95.88	96.88
Right kidney	DSC	85.34	84.13	92.23	90.44	93.79	93.32	94.22
	Precision	87.83	84.23	95.88	91.67	94.67	95.11	95.34
	Recall	87.03	88.46	91.19	92.57	94.65	94.45	94.61