TABLE 6.

Performance comparison with SOTA methods on ACDC datasets. $\text{Red}$ , $\text{Green}$ , and $\text{Blue}$ indicate the best, second-best, and third-best performance.

Network	ACDC datasets				#Params (M)	Model size (MB)
	RV	Myo	LV	Average
U-Net Ronneberger et al. (2015)	0.743 (0.792)	0.717 (0.812)	0.861 (0.897)	0.774 (0.834)	29.59	118
SegNet Badrinarayanan et al. (2017)	0.738 (0.790)	0.720 (0.817)	$\text{0.864 (0.902)}$	0.774 (0.836)	17.94	71.8
FATNet Wu et al. (2022)	0.743 (0.799)	0.702 (0.805)	0.859 (0.899)	0.768 (0.834)	27.43	109
Swin-UNet Cao et al. (2021)	$\text{0.754 (0.805)}$	$\text{0.722 (0.820)}$	$\text{0.865 (0.903)}$	$\text{0.780 (0.843)}$	25.86	105
TransUNet Chen et al. (2021b)	$\text{0.750 (0.800)}$	0.715 (0.812)	$\text{0.872 (0.905)}$	$\text{0.779 (0.839)}$	88.87	401
EANet Wang et al. (2022)	0.742 (0.791)	$\text{0.732 (0.825)}$	0.864 (0.902)	$\text{0.779 (0.839)}$	47.07	118
UNeXt-L Valanarasu and Patel, (2022)	0.719 (0.779)	0.675 (0.810)	0.840 (0.882)	0.745 (0.824)	$\text{3.99}$	$\text{15.2}$
PL-Net† (Our)	0.723 (0.778)	0.692 (0.796)	0.845 (0.887)	0.753 (0.820)	$\text{3.77}$	$\text{15.6}$
PL-Net (Our)	$\text{0.761 (0.807)}$	$\text{0.738 (0.828)}$	$\text{0.872 (0.907)}$	$\text{0.790 (0.847)}$	$\text{15.03}$	$\text{60.7}$

The bold values indicates the best performance.