Research on CT Lung Segmentation Method of Preschool Children based on Traditional Image Processing and ResUnet

Table 3

Segmentation results of different segmentation algorithms.

Method	IOU	Dice	Precision	Recall
Unet	0.934	0.9553	0.9561	0.9625
Unet++	0.9338	0.9543	0.9493	0.9699
Unet+++	0.9359	0.9568	0.9468	0.9748
Attention-UNet	0.9257	0.9482	0.9424	0.9673
Swin-Unet	0.8738	0.917	0.8998	0.949
Trans-Unet	0.9364	0.9569	0.9611	0.9592
Proposed method	0.9479	0.9678	0.9711	0.9715
ResUnet (adult)	0.9163	0.9368	0.9351	0.9451