Table 3.

Performance comparison of the proposed multimodal auto-context fused IOUC-3DSFCNN model with 3DSFCNN model.

Methods	DICE			Recall			Precision			HD(mm)
	Whole	Core	Enh.	Whole	Core	Enh.	Whole	Core	Enh.	Whole	Core	Enh.
IOUC-3DSFCNN	0.82	0.81	0.70	0.88	0.80	0.79	0.84	0.86	0.74	7.63	8.31	9.08
IOUC-3DSFCNN + Auto-context	0.89	0.84	0.78	0.91	0.84	0.85	0.88	0.85	0.78	4.03	6.32	7.85