TY - JOUR
T1 - QS-ADN
T2 - quasi-supervised artifact disentanglement network for low-dose CT image denoising by local similarity among unpaired data
AU - Ruan, Yuhui
AU - Yuan, Qiao
AU - Niu, Chuang
AU - Li, Chen
AU - Yao, Yudong
AU - Wang, Ge
AU - Teng, Yueyang
N1 - Publisher Copyright:
© 2023 Institute of Physics and Engineering in Medicine
PY - 2023/10/21
Y1 - 2023/10/21
N2 - Deep learning has been successfully applied to low-dose CT (LDCT) image denoising for reducing potential radiation risk. However, the widely reported supervised LDCT denoising networks require a training set of paired images, which is expensive to obtain and cannot be perfectly simulated. Unsupervised learning utilizes unpaired data and is highly desirable for LDCT denoising. As an example, an artifact disentanglement network (ADN) relies on unpaired images and obviates the need for supervision but the results of artifact reduction are not as good as those through supervised learning. An important observation is that there is often hidden similarity among unpaired data that can be utilized. This paper introduces a new learning mode, called quasi-supervised learning, to empower ADN for LDCT image denoising. For every LDCT image, the best matched image is first found from an unpaired normal-dose CT (NDCT) dataset. Then, the matched pairs and the corresponding matching degree as prior information are used to construct and train our ADN-type network for LDCT denoising. The proposed method is different from (but compatible with) supervised and semi-supervised learning modes and can be easily implemented by modifying existing networks. The experimental results show that the method is competitive with state-of-the-art methods in terms of noise suppression and contextual fidelity. The code and working dataset are publicly available at https://github.com/ruanyuhui/ADN-QSDL.git.
AB - Deep learning has been successfully applied to low-dose CT (LDCT) image denoising for reducing potential radiation risk. However, the widely reported supervised LDCT denoising networks require a training set of paired images, which is expensive to obtain and cannot be perfectly simulated. Unsupervised learning utilizes unpaired data and is highly desirable for LDCT denoising. As an example, an artifact disentanglement network (ADN) relies on unpaired images and obviates the need for supervision but the results of artifact reduction are not as good as those through supervised learning. An important observation is that there is often hidden similarity among unpaired data that can be utilized. This paper introduces a new learning mode, called quasi-supervised learning, to empower ADN for LDCT image denoising. For every LDCT image, the best matched image is first found from an unpaired normal-dose CT (NDCT) dataset. Then, the matched pairs and the corresponding matching degree as prior information are used to construct and train our ADN-type network for LDCT denoising. The proposed method is different from (but compatible with) supervised and semi-supervised learning modes and can be easily implemented by modifying existing networks. The experimental results show that the method is competitive with state-of-the-art methods in terms of noise suppression and contextual fidelity. The code and working dataset are publicly available at https://github.com/ruanyuhui/ADN-QSDL.git.
KW - CT image denoising
KW - deep learning
KW - local similarity
KW - quasi-supervised learning
KW - unpaired data
UR - http://www.scopus.com/inward/record.url?scp=85175307436&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85175307436&partnerID=8YFLogxK
U2 - 10.1088/1361-6560/acf9da
DO - 10.1088/1361-6560/acf9da
M3 - Article
AN - SCOPUS:85175307436
SN - 0031-9155
VL - 68
JO - Physics in Medicine and Biology
JF - Physics in Medicine and Biology
IS - 20
M1 - 205001
ER -