Skip to main content
. 2021 Nov 23;8:768245. doi: 10.3389/fcvm.2021.768245

Table 1.

Human observer assessment of motion extent (%) on 200 T1 maps before motion correction, and after the baseline and proposed method (MOCOnet) for motion correction.

All data (N = 200) Group 1 (N = 50) Severe motion Group 2 (N = 100) Moderate motion Group 3 (N = 50) Mild to no motion
Before MOCO 37.1 ± 21.5 55.8 ± 18.7 (99.3) 35.5 ± 18.9 (80.5) 21.7 ± 13.8 (62.1)
Baseline method 15.8 ± 15.6 25.8 ± 19.8 (93.4) 14.7 ± 13.9 (65.7) 8.1 ± 6.5 (34.2)
MOCOnet 13.3 ± 10.5 18.6 ± 14.3 (86.9) 12.7 ± 9.2 (46.4) 9.4 ± 6.4 (19.8)

The quality scores are inverse variance-weighted scores of three human observers and reported in mean ± SD (maximum value). The best results are highlighted in bold.