Table 5.

Performance comparison of segmentation models built upon scans of 4 imaging modalities and 3 imaging modalities (the sizes of patches used to train FCNNs were 33^*33^*4 and 65^*65^*4, or 33^*33^*3 and 65^*65^*3 respectively, n = 5, and the number of patches used to train FCNNs was 5000^*5^*20).

Dataset	No. of modalities	Methods	Dice			PPV			Sensitivity
			Complete	Core	Enhancing	Complete	Core	Enhancing	Complete	Core	Enhancing
Challenge	4	FCNNs	0.74	0.72	0.67	0.61	0.64	0.60	0.95	0.86	0.77
		FCNN + CRF	0.85	0.80	0.71	0.85	0.79	0.67	0.87	0.82	0.78
		FCNN + CRF + post-process	0.87	0.83	0.76	0.91	0.85	0.78	0.84	0.82	0.76
	3	FCNNs	0.74	0.72	0.67	0.62	0.63	0.60	0.94	0.86	0.77
		FCNN + CRF	0.85	0.80	0.70	0.87	0.80	0.63	0.84	0.81	0.80
		FCNN + CRF + post-process	0.87	0.83	0.76	0.92	0.87	0.77	0.83	0.81	0.77
Learderboard	4	FCNNs	0.69	0.59	0.55	0.57	0.56	0.49	0.95	0.72	0.68
		FCNN + CRF	0.82	0.64	0.58	0.82	0.70	0.55	0.85	0.67	0.68
		FCNN + CRF + post-process	0.85	0.74	0.62	0.89	0.75	0.62	0.83	0.79	0.66
	3	FCNNs	0.70	0.61	0.54	0.58	0.57	0.49	0.96	0.74	0.67
		FCNN + CRF	0.83	0.66	0.57	0.85	0.71	0.50	0.85	0.69	0.71
		FCNN + CRF + post-process	0.86	0.73	0.62	0.89	0.76	0.64	0.84	0.78	0.68