Table 3.

Accuracy of DeepLab V3+ image segmentation models.

	DICE loss			IoU
	Train set	Validation set	Test set	Train set	Validation set	Test set
Foot	0.0110	0.0110	0.0123	0.9938	0.9931	0.9884
First metatarsal	0.0214	0.0222	0.0241	0.9574	0.9567	0.9513
Talus	0.0297	0.0272	0.0302	0.9511	0.9589	0.9487
Calcaneus	0.0159	0.0153	0.0159	0.9728	0.9742	0.9731
Navicular	0.0341	0.0340	0.0361	0.9483	0.9472	0.9446