Reference-Driven Undersampled MR Image Reconstruction Using Wavelet Sparsity-Constrained Deep Image Prior

Table 1

Parameter settings for experiments.

Parameter		Images
		Brain A	Brain B	Brain C
Network hyperparameters	Learning rate	0.01	0.01	0.01
		6	6	6
		[16, 32, 64, 64, 128, 128]	[32, 32, 64, 128, 128, 128]	[32, 32, 64, 128, 128, 128]
		[16, 32, 64, 64, 128, 128]	[32, 32, 64, 128, 128, 128]	[32, 32, 64, 128, 128, 128]
		[16,16,16,16,16,16]	[16,16,16,16,16,16]	[16,16,16,16,16,16]
		[3,3,3,3,3,3]	[3,3,3,3,3,3]	[3,3,3,3,3,3]
		[3,3,3,3,3,3]	[3,3,3,3,3,3]	[3,3,3,3,3,3]
		[1,1,1,1,1,1]	[1,1,1,1,1,1]	[1,1,1,1,1,1]
Iteration number	MaxIt	50	50	50
	SubIt	100	100	100
Wavelet parameters	Wavelet function	Haar	Haar	Haar
	Decomposition level	8	6	6
		0.07	0.05	0.05
		0.0001	0.0001	0.0001