BMC Medical Imaging

Table 5 SSIM values of test images

From: Artifact estimation network for MR images: effectiveness of batch normalization and dropout layers

	1	2	3	4	5	6	7	8	9	10
Artifact	0.889	0.892	0.869	0.856	0.866	0.856	0.871	0.889	0.896	0.898
Transformer	0.900	0.890	0.894	0.897	0.887	0.894	0.895	0.897	0.903	0.873
Network 1	0.947	0.951	0.936	0.925	0.94	0.924	0.942	0.954	0.958	0.956
Network 2	0.974	0.974	0.974	0.967	0.974	0.967	0.976	0.978	0.978	0.979
Network 3	0.973	0.974	0.973	0.963	0.973	0.966	0.975	0.978	0.978	0.977

SSIM values measured using 10 test images. The numbers listed are the averages of five training and testing runs

Back to article page

ISSN: 1471-2342

Contact us

General enquiries: journalsubmissions@springernature.com