Skip to main content
. Author manuscript; available in PMC: 2023 Jun 1.
Published in final edited form as: IEEE Trans Med Imaging. 2022 Jun 1;41(6):1331–1345. doi: 10.1109/TMI.2021.3139999

TABLE VI:

Results of a fully-supervised 3D UNet trained with Shadow-AUG and Shadow-DROP mechanisms using hard/soft shadow thresholding functions. Bold values indicate the best result. Underlined values suggest results without statistical significance compared with the bottom row (p>0.05).

Models UCLA dataset NIH dataset
DSC[%] ASD[mm] DSC[%] ASD[mm]
Hard threshold 92.24 (2.19) 0.93 (0.30) 89.82 (3.57) 1.28 (0.63)
Soft threshold 92.25 (2.19) 0.93 (0.29) 89.85 (3.30) 1.26 (0.58)