Improved Complex Convolutional Neural Network Based on SPIRiT and Dense Connection for Parallel MRI Reconstruction

Table 6

Comparisons of average PSNR and SSIM values with standard deviation of the MR images reconstructed by SPIRiT, Deepcomplex, Deepcomplex with SCB, and our proposed SPIRiT-Net on the Coronal-PD dataset under the 2DPU pattern.

Reconstruction model	Threefold AF		Fivefold AF
	PSNR (dB)	SSIM	PSNR (dB)	SSIM
SPIRiT	39.398 ± 2.762	0.928 ± 0.026	36.832 ± 2.450	0.883 ± 0.034
Deepcomplex	39.491 ± 2.704	0.934 ± 0.022	37.151 ± 2.450	0.894 ± 0.034
Deepcomplex with SCB	40.551 ± 1.591	0.940 ± 0.012	38.595 ± 1.515	0.916 ± 0.015
SPIRiT-Net	40.869 ± 1.702	0.942 ± 0.016	38.941 ± 1.606	0.919 ± 0.019

The best values are in bold.