PMID- 32276185
OWN - NLM
STAT- MEDLINE
DCOM- 20210531
LR  - 20210531
IS  - 1361-8423 (Electronic)
IS  - 1361-8415 (Linking)
VI  - 62
DP  - 2020 May
TI  - Contrast agent-free synthesis and segmentation of ischemic heart disease images 
      using progressive sequential causal GANs.
PG  - 101668
LID - S1361-8415(20)30034-7 [pii]
LID - 10.1016/j.media.2020.101668 [doi]
AB  - The elimination of gadolinium contrast agent (CA) injections and manual 
      segmentation are crucial for ischemic heart disease (IHD) diagnosis and 
      treatment. In the clinic, CA-based late gadolinium enhancement (LGE) imaging and 
      manual segmentation remain subject to concerns about potential toxicity, 
      interobserver variability, and ineffectiveness. In this study, progressive 
      sequential causal GANs (PSCGAN) are proposed. This is the first one-stop CA-free 
      IHD technology that can simultaneously synthesize an LGE-equivalent image and 
      segment diagnosis-related tissues (i.e., scars, healthy myocardium, blood pools, 
      and other pixels) from cine MR images. To this end, the PSCGAN offer three unique 
      properties: 1) a progressive framework that cascades three phases (i.e., priori 
      generation, conditional synthesis, and enhanced segmentation) for 
      divide-and-conquer training synthesis and segmentation of images. Importantly, 
      this framework leverages the output of the previous phase as a priori condition 
      to input the next phase and guides its training for enhancing performance, 2) a 
      sequential causal learning network (SCLN) that creates a multi-scale, two-stream 
      pathway and a multi-attention weighing unit to extract spatial and temporal 
      dependencies from cine MR images and effectively select task-specific dependence. 
      It also integrates the GAN architecture to leverage adversarial training to 
      further facilitate the learning of interest dependencies of the latent space of 
      cine MR images in all phases; and 3) two specifically designed self-learning loss 
      terms: a synthetic regularization loss term leverages the spare regularization to 
      avoid noise during synthesis, and a segmentation auxiliary loss term leverages 
      the number of pixels for each tissue to compensate for discrimination during 
      segmentation. Thus, the PSCGAN gain unprecedented performance while stably 
      training in both synthesis and segmentation. By training and testing a total of 
      280 clinical subjects, our PSCGAN yield a synthetic normalization 
      root-mean-squared-error of 0.14 and an overall segmentation accuracy of 97.17%. 
      It also produces a 0.96 correlation coefficient for the scar ratio in a real 
      diagnostic metric evaluation. These results proved that our method is able to 
      offer significant assistance in the standardized assessment of cardiac disease.
CI  - Copyright (c) 2020 Elsevier B.V. All rights reserved.
FAU - Xu, Chenchu
AU  - Xu C
AD  - Department of Medical Imaging, Western University, London ON, Canada.
FAU - Xu, Lei
AU  - Xu L
AD  - Department of Radiology, Beijing AnZhen Hospital, Beijing, China.
FAU - Ohorodnyk, Pavlo
AU  - Ohorodnyk P
AD  - Department of Medical Imaging, Western University, London ON, Canada.
FAU - Roth, Mike
AU  - Roth M
AD  - Department of Medical Imaging, Western University, London ON, Canada.
FAU - Chen, Bo
AU  - Chen B
AD  - School of Health Science, Western University, London ON, Canada.
FAU - Li, Shuo
AU  - Li S
AD  - Department of Medical Imaging, Western University, London ON, Canada. Electronic 
      address: sli287@uwo.ca.
LA  - eng
PT  - Journal Article
DEP - 20200226
PL  - Netherlands
TA  - Med Image Anal
JT  - Medical image analysis
JID - 9713490
RN  - 0 (Contrast Media)
RN  - AU0V1LM3JT (Gadolinium)
SB  - IM
MH  - *Contrast Media
MH  - Gadolinium
MH  - Humans
MH  - *Myocardial Ischemia/diagnostic imaging
MH  - Myocardium
MH  - Observer Variation
OTO - NOTNLM
OT  - Gadolinium contrast agents
OT  - Ischemic heart disease
OT  - Progressive framework
OT  - Sequential learning
OT  - Synthesis
COIS- Declaration of Competing Interest None.
EDAT- 2020/04/11 06:00
MHDA- 2021/06/01 06:00
CRDT- 2020/04/11 06:00
PHST- 2019/09/27 00:00 [received]
PHST- 2020/01/17 00:00 [revised]
PHST- 2020/02/21 00:00 [accepted]
PHST- 2020/04/11 06:00 [pubmed]
PHST- 2021/06/01 06:00 [medline]
PHST- 2020/04/11 06:00 [entrez]
AID - S1361-8415(20)30034-7 [pii]
AID - 10.1016/j.media.2020.101668 [doi]
PST - ppublish
SO  - Med Image Anal. 2020 May;62:101668. doi: 10.1016/j.media.2020.101668. Epub 2020 
      Feb 26.