Table 1.

Thoracic wall segmentation performance of each set of combinations of MF and CY with the two CNNs.

CNN	MF	CY	mIoU		mDice		mPrecision		mRecall
			Mean	SD	Mean	SD	Mean	SD	Mean	SD
U-net			0.448	0.017	0.610	0.016	0.679	0.033	0.568	0.032
	✓		0.461	0.017	0.623	0.016	0.659	0.035	0.606	0.032
		✓	0.486	0.006	0.647	0.005	0.599	0.014	0.718	0.025
	✓	✓	0.493	0.006	0.654	0.005	0.596	0.016	0.738	0.023
DeepLabv3+			0.417	0.007	0.582	0.007	0.675	0.009	0.525	0.013
	✓		0.443	0.006	0.608	0.006	0.663	0.009	0.575	0.014
		✓	0.462	0.005	0.626	0.005	0.567	0.004	0.709	0.012
	✓	✓	0.470	0.004	0.633	0.004	0.566	0.007	0.729	0.010

mIoU: the mean value of intersection over union; mDice: the mean value of Dice; mPrecision: the mean value of Precision; mRecall: the mean value of Recall; SD: standard deviation.