A Network with Composite Loss and Parameter-free Chunking Fusion Block for Super-Resolution MR Image

Table 3

Performance and parameter comparison of the proposed module. The super-resolution reconstructed LR image is obtained by BD.

	Residual block	Reconstruction block	PCFB		PSNR	SSIM	RMSE	Params (MB)
SRResnet [51]	✓	✓	—	PD	34.32	0.9783	0.0203	5.28
				T1	32.90	0.9636	0.0234
				T2	35.47	0.9761	0.0174
Proposed	✓	✓	✓	PD	34.52	0.9790	0.0200	5.27
				T1	33.23	0.9659	0.0227
				T2	35.53	0.9761	0.0173

Bold values highlight the optimal results.