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

Table 4

Performance of the Unet model applied to adult lung segmentation on our test set.

Case no.	IOU	Dice	Precision	Recall
1	0.9226	0.9436	0.9357	0.9535
2	0.9305	0.9497	0.9481	0.9514
3	0.9164	0.9397	0.937	0.9434
4	0.9291	0.9477	0.9413	0.9558
5	0.9248	0.9441	0.9438	0.9494
6	0.9172	0.9391	0.9283	0.9562
7	0.9261	0.9459	0.9436	0.9505
8	0.9295	0.9483	0.9507	0.9476
9	0.8561	0.8766	0.8764	0.909
10	0.912	0.9333	0.9365	0.9415
11	0.9158	0.9394	0.9331	0.9497
12	0.9151	0.9345	0.9462	0.9329
Average	0.9163	0.9368	0.9351	0.9451