Table 2. Quantitative results of different algorithms for all 526 testing images.

Algorithm	PSNR	SSIM	NRMSE
FBP	33.8273±2.0114	0.7475±0.0774	0.0209±0.0049
TV	39.5959±1.7598	0.9261±0.0245	0.0107±0.0021
RedCNN-MSE	42.7613±1.8630	0.9626±0.0137	0.0074±0.0016
ADAPTIVE-MSE	41.7831±1.8240	0.9639±0.0140	0.0083±0.0016
RedCNN-VGG-1000*MSE	43.9774±2.0509	0.9685±0.0131	0.0065±0.0016
ADAPTIVE-VGG-5000*MSE	43.1503±1.9297	0.9680±0.0131	0.0071±0.0016
ADAPTIVE-VGG-1000*MSE	42.5767±1.9243	0.9652±0.0138	0.0076±0.0017

FBP, filtered-back-projection; NRMSE, normalized root mean square error; SSIM, structural similarity index metric; PSNR, peak signal-to-noise ratio.