Characterization of interstitial diffuse fibrosis patterns using texture analysis of myocardial native T1 mapping.
Adult
Aged
Cardiomyopathy, Dilated
/ diagnostic imaging
Cardiomyopathy, Hypertrophic
/ diagnostic imaging
Endomyocardial Fibrosis
/ diagnostic imaging
Female
Humans
Image Interpretation, Computer-Assisted
/ methods
Magnetic Resonance Imaging
/ methods
Male
Middle Aged
Sensitivity and Specificity
Support Vector Machine
Journal
PloS one
ISSN: 1932-6203
Titre abrégé: PLoS One
Pays: United States
ID NLM: 101285081
Informations de publication
Date de publication:
2020
2020
Historique:
received:
03
09
2019
accepted:
11
05
2020
entrez:
2
6
2020
pubmed:
2
6
2020
medline:
20
8
2020
Statut:
epublish
Résumé
The pattern of myocardial fibrosis differs significantly between different cardiomyopathies. Fibrosis in hypertrophic cardiomyopathy (HCM) is characteristically as patchy and regional but in dilated cardiomyopathy (DCM) as diffuse and global. We sought to investigate if texture analyses on myocardial native T1 mapping can differentiate between fibrosis patterns in patients with HCM and DCM. We prospectively acquired native myocardial T1 mapping images for 321 subjects (55±15 years, 70% male): 65 control, 116 HCM, and 140 DCM patients. To quantify different fibrosis patterns, four sets of texture descriptors were used to extract 152 texture features from native T1 maps. Seven features were sequentially selected to identify HCM- and DCM-specific patterns in 70% of data (training dataset). Pattern reproducibility and generalizability were tested on the rest of data (testing dataset) using support vector machines (SVM) and regression models. Pattern-derived texture features were capable to identify subjects in HCM, DCM, and controls cohorts with 202/237(85.2%) accuracy of all subjects in the training dataset using 10-fold cross-validation on SVM (AUC = 0.93, 0.93, and 0.93 for controls, HCM and DCM, respectively), while pattern-independent global native T1 mapping was poorly capable to identify those subjects with 121/237(51.1%) accuracy (AUC = 0.78, 0.51, and 0.74) (P<0.001 for all). The pattern-derived features were reproducible with excellent intra- and inter-observer reliability and generalizable on the testing dataset with 75/84(89.3%) accuracy. Texture analysis of myocardial native T1 mapping can characterize fibrosis patterns in HCM and DCM patients and provides additional information beyond average native T1 values.
Sections du résumé
BACKGROUND
The pattern of myocardial fibrosis differs significantly between different cardiomyopathies. Fibrosis in hypertrophic cardiomyopathy (HCM) is characteristically as patchy and regional but in dilated cardiomyopathy (DCM) as diffuse and global. We sought to investigate if texture analyses on myocardial native T1 mapping can differentiate between fibrosis patterns in patients with HCM and DCM.
METHODS
We prospectively acquired native myocardial T1 mapping images for 321 subjects (55±15 years, 70% male): 65 control, 116 HCM, and 140 DCM patients. To quantify different fibrosis patterns, four sets of texture descriptors were used to extract 152 texture features from native T1 maps. Seven features were sequentially selected to identify HCM- and DCM-specific patterns in 70% of data (training dataset). Pattern reproducibility and generalizability were tested on the rest of data (testing dataset) using support vector machines (SVM) and regression models.
RESULTS
Pattern-derived texture features were capable to identify subjects in HCM, DCM, and controls cohorts with 202/237(85.2%) accuracy of all subjects in the training dataset using 10-fold cross-validation on SVM (AUC = 0.93, 0.93, and 0.93 for controls, HCM and DCM, respectively), while pattern-independent global native T1 mapping was poorly capable to identify those subjects with 121/237(51.1%) accuracy (AUC = 0.78, 0.51, and 0.74) (P<0.001 for all). The pattern-derived features were reproducible with excellent intra- and inter-observer reliability and generalizable on the testing dataset with 75/84(89.3%) accuracy.
CONCLUSION
Texture analysis of myocardial native T1 mapping can characterize fibrosis patterns in HCM and DCM patients and provides additional information beyond average native T1 values.
Identifiants
pubmed: 32479518
doi: 10.1371/journal.pone.0233694
pii: PONE-D-19-24820
pmc: PMC7263579
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
e0233694Subventions
Organisme : NHLBI NIH HHS
ID : R01 HL129157
Pays : United States
Organisme : NHLBI NIH HHS
ID : R01 HL127015
Pays : United States
Organisme : NHLBI NIH HHS
ID : R01 HL129185
Pays : United States
Déclaration de conflit d'intérêts
The authors have declared that no competing interests exist.
Références
J Cardiovasc Magn Reson. 2017 Oct 9;19(1):75
pubmed: 28992817
Circ Cardiovasc Imaging. 2015 Dec;8(12):
pubmed: 26659373
J Cardiovasc Magn Reson. 2017 Sep 28;19(1):74
pubmed: 28954631
Circ Cardiovasc Imaging. 2012 Nov;5(6):726-33
pubmed: 23071146
Medicine (Baltimore). 2018 Sep;97(37):e12246
pubmed: 30212958
J Cardiovasc Magn Reson. 2015 Apr 18;17:29
pubmed: 25928314
JACC Cardiovasc Imaging. 2016 Jan;9(1):40-50
pubmed: 26762873
Eur J Radiol. 2018 May;102:61-67
pubmed: 29685546
PLoS One. 2016 May 24;11(5):e0155591
pubmed: 27219329
J Am Soc Echocardiogr. 2014 Sep;27(9):911-39
pubmed: 25172399
JACC Cardiovasc Imaging. 2018 Dec;11(12):1917-1918
pubmed: 30121270
Biometrics. 1988 Sep;44(3):837-45
pubmed: 3203132
Eur Heart J Cardiovasc Imaging. 2015 Jan;16(1):14-22
pubmed: 25354866
Radiology. 2016 Feb;278(2):563-77
pubmed: 26579733
IEEE Trans Image Process. 1998;7(11):1602-9
pubmed: 18276225
J Cardiovasc Magn Reson. 2016 Nov 16;18(1):79
pubmed: 27846845
Eur Heart J. 2008 Jan;29(2):270-6
pubmed: 17916581
Radiology. 2018 Jan;286(1):103-112
pubmed: 28836886
Am J Cardiol. 2016 Oct 1;118(7):1057-62
pubmed: 27567135
JACC Cardiovasc Imaging. 2013 Apr;6(4):475-84
pubmed: 23498674
JACC Cardiovasc Imaging. 2018 Jan;11(1):48-59
pubmed: 28624408
Circulation. 2010 Jul 13;122(2):138-44
pubmed: 20585010
Magn Reson Med. 2015 Apr;73(4):1469-82
pubmed: 24798588
J Cardiovasc Magn Reson. 2013 Apr 10;15:31
pubmed: 23574733
J Am Coll Cardiol. 2002 Mar 20;39(6):1055-60
pubmed: 11897450
Am J Cardiol. 2016 Jan 15;117(2):282-8
pubmed: 26684511
Eur Heart J. 2014 Oct 14;35(39):2733-79
pubmed: 25173338
J Magn Reson Imaging. 2018 Mar;47(3):779-786
pubmed: 28737018
Magn Reson Med. 2018 Aug;80(2):780-791
pubmed: 29314198
Sci Rep. 2016 Jun 16;6:27988
pubmed: 27306927
Circ Cardiovasc Imaging. 2009 May;2(3):183-90
pubmed: 19808591
Eur Heart J Cardiovasc Imaging. 2018 Jul 1;19(7):768-776
pubmed: 29237044
JACC Cardiovasc Imaging. 2019 Oct;12(10):1946-1954
pubmed: 30660549
Radiology. 2018 Nov;289(2):357-365
pubmed: 30084736