PMID- 29994362
OWN - NLM
STAT- MEDLINE
DCOM- 20190624
LR  - 20190802
IS  - 1558-254X (Electronic)
IS  - 0278-0062 (Print)
IS  - 0278-0062 (Linking)
VI  - 37
IP  - 8
DP  - 2018 Aug
TI  - Deep Learning for Quantification of Epicardial and Thoracic Adipose Tissue From 
      Non-Contrast CT.
PG  - 1835-1846
LID - 10.1109/TMI.2018.2804799 [doi]
AB  - Epicardial adipose tissue (EAT) is a visceral fat deposit related to coronary 
      artery disease. Fully automated quantification of EAT volume in clinical routine 
      could be a timesaving and reliable tool for cardiovascular risk assessment. We 
      propose a new fully automated deep learning framework for EAT and thoracic 
      adipose tissue (TAT) quantification from non-contrast coronary artery calcium 
      computed tomography (CT) scans. The first multi-task convolutional neural network 
      (ConvNet) is used to determine heart limits and perform segmentation of heart and 
      adipose tissues. The second ConvNet, combined with a statistical shape model, 
      allows for pericardium detection. EAT and TAT segmentations are then obtained 
      from outputs of both ConvNets. We evaluate the performance of the method on CT 
      data sets from 250 asymptomatic individuals. Strong agreement between automatic 
      and expert manual quantification is obtained for both EAT and TAT with median 
      Dice score coefficients of 0.823 (inter-quartile range (IQR): 0.779-0.860) and 
      0.905 (IQR: 0.862-0.928), respectively; with excellent correlations of 0.924 and 
      0.945 for EAT and TAT volumes. Computations are performed in <6 s on a standard 
      personal computer for one CT scan. Therefore, the proposed method represents a 
      tool for rapid fully automated quantification of adipose tissue and may improve 
      cardiovascular risk stratification in patients referred for routine CT calcium 
      scans.
FAU - Commandeur, Frederic
AU  - Commandeur F
FAU - Goeller, Markus
AU  - Goeller M
FAU - Betancur, Julian
AU  - Betancur J
FAU - Cadet, Sebastien
AU  - Cadet S
FAU - Doris, Mhairi
AU  - Doris M
FAU - Chen, Xi
AU  - Chen X
FAU - Berman, Daniel S
AU  - Berman DS
FAU - Slomka, Piotr J
AU  - Slomka PJ
FAU - Tamarappoo, Balaji K
AU  - Tamarappoo BK
FAU - Dey, Damini
AU  - Dey D
LA  - eng
GR  - R01 HL133616/HL/NHLBI NIH HHS/United States
PT  - Journal Article
PT  - Research Support, N.I.H., Extramural
DEP - 20180209
PL  - United States
TA  - IEEE Trans Med Imaging
JT  - IEEE transactions on medical imaging
JID - 8310780
SB  - IM
MH  - Adipose Tissue/*diagnostic imaging
MH  - Aged
MH  - *Deep Learning
MH  - Female
MH  - Humans
MH  - Image Processing, Computer-Assisted/*methods
MH  - Male
MH  - Middle Aged
MH  - Pericardium/*diagnostic imaging
MH  - Thorax/diagnostic imaging
MH  - Tomography, X-Ray Computed/*methods
PMC - PMC6076348
MID - NIHMS943682
EDAT- 2018/07/12 06:00
MHDA- 2019/06/25 06:00
PMCR- 2019/08/01
CRDT- 2018/07/12 06:00
PHST- 2018/07/12 06:00 [pubmed]
PHST- 2019/06/25 06:00 [medline]
PHST- 2018/07/12 06:00 [entrez]
PHST- 2019/08/01 00:00 [pmc-release]
AID - 10.1109/TMI.2018.2804799 [doi]
PST - ppublish
SO  - IEEE Trans Med Imaging. 2018 Aug;37(8):1835-1846. doi: 10.1109/TMI.2018.2804799. 
      Epub 2018 Feb 9.