Table 2. Quantitative comparisons with respect to ground truth for different dense layers included the UNET (Ronneberger, Fischer & Brox, 2015) and Res-Net (He et al., 2016) architecture.

ACC, DC, Sen, Sp, Pc, AUC, F1, Sm, Ea and MAE represent accuracy, Dice coefficient, sensitivity, specificity, precision, the area under the curve, F1 score, structural metric, enhancement alignment meter and mean absolute error, respectively.

Number of dense network	ACC	DC	Sen	Sp	Pc	AUC	F1	Sm	Eα	MAE
Num1	0.9696	0.7971	0.8011	0.9958	0.8290	0.9513	0.8129	0.8411	0.9315	0.0088
Num2	0.9700	0.8011	0.8096	0.9966	0.8596	0.9492	0.8184	0.8528	0.9394	0.0083
Num3	0.9686	0.7569	0.7546	0.9957	0.8200	0.9334	0.7806	0.8349	0.9379	0.0104
Num4	0.9699	0.7869	0.7579	0.9961	0.8485	0.9495	0.8241	0.8341	0.9348	0.0090
UNET	0.9696	0.7998	0.8052	0.9957	0.8247	0.9347	0.8154	0.8400	0.9390	0.0088
Res-Net	0.9698	0.8002	0.7978	0.9962	0.8344	0.9504	0.8180	0.8415	0.9352	0.0094