Title
Supervised learning with cyclegan for low-dose FDG PET image denoising.
Abstract
•Low-dose PET image denoising or reconstruction.•Designed a cycle wasserstein regression adversarial training framework.•Evaluate the maximum and mean standardized uptake value (SUV) deviation on real clinic PET image.•The bias of mean and maximum SUV for lesions and normal tissues are less than 5% and 2%, respectively.
Year
DOI
Venue
2020
10.1016/j.media.2020.101770
Medical Image Analysis
Keywords
DocType
Volume
PET,Low-dose,Generative adversarial networks,Cycle consistent
Journal
65
ISSN
Citations 
PageRank 
1361-8415
2
0.49
References 
Authors
0
5
Name
Order
Citations
PageRank
Long Zhou12911.00
Joshua D Schaefferkoetter220.49
I W K Tham341.25
Gang Huang464.38
Jianhua Yan520.49