Open AccessPET image denoising using unsupervised deep learning
Author(s) -
Juan Cui,
Kuang Gong,
Ning Guo,
Chenxi Wu,
Xiaojuan Meng,
Kyungsang Kim,
Kun Zheng,
Zhiyong Wu,
Liping Fu,
Baixuan Xu,
Z.H. Zhu,
Jiahe Tian,
Huafeng Liu,
Quanzheng Li
Publication year - 2019
Publication title -
european journal of nuclear medicine and molecular imaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.313
H-Index - 163
eISSN - 1619-7089
pISSN - 1619-7070
DOI - 10.1007/s00259-019-04468-4
Subject(s) - wilcoxon signed rank test , artificial intelligence , imaging phantom , image quality , computer science , pattern recognition (psychology) , deep learning , gaussian , noise reduction , standard deviation , image (mathematics) , nuclear medicine , mathematics , medicine , statistics , physics , mann–whitney u test , quantum mechanics
Image quality of positron emission tomography (PET) is limited by various physical degradation factors. Our study aims to perform PET image denoising by utilizing prior information from the same patient. The proposed method is based on unsupervised deep learning, where no training pairs are needed.