PMID- 31714190
OWN - NLM
STAT- MEDLINE
DCOM- 20200420
LR  - 20240214
IS  - 1527-1315 (Electronic)
IS  - 0033-8419 (Print)
IS  - 0033-8419 (Linking)
VI  - 294
IP  - 1
DP  - 2020 Jan
TI  - Three-dimensional Deep Convolutional Neural Networks for Automated Myocardial 
      Scar Quantification in Hypertrophic Cardiomyopathy: A Multicenter Multivendor 
      Study.
PG  - 52-60
LID - 10.1148/radiol.2019190737 [doi]
AB  - Background Cardiac MRI late gadolinium enhancement (LGE) scar volume is an 
      important marker for outcome prediction in patients with hypertrophic 
      cardiomyopathy (HCM); however, its clinical application is hindered by a lack of 
      measurement standardization. Purpose To develop and evaluate a three-dimensional 
      (3D) convolutional neural network (CNN)-based method for automated LGE scar 
      quantification in patients with HCM. Materials and Methods We retrospectively 
      identified LGE MRI data in a multicenter (n = 7) and multivendor (n = 3) HCM 
      study obtained between November 2001 and November 2011. A deep 3D CNN based on 
      U-Net architecture was used for LGE scar quantification. Independent CNN training 
      and testing data sets were maintained with a 4:1 ratio. Stacks of short-axis MRI 
      slices were split into overlapping substacks that were segmented and then merged 
      into one volume. The 3D CNN per-site and per-vendor performances were evaluated 
      with respect to manual scar quantification performed in a core laboratory setting 
      using Dice similarity coefficient (DSC), Pearson correlation, and Bland-Altman 
      analyses. Furthermore, the performance of 3D CNN was compared with that of 
      two-dimensional (2D) CNN. Results This study included 1073 patients with HCM (733 
      men; mean age, 49 years +/- 17 [standard deviation]). The 3D CNN-based 
      quantification was fast (0.15 second per image) and demonstrated excellent 
      correlation with manual scar volume quantification (r = 0.88, P < .001) and ratio 
      of scar volume to total left ventricle myocardial volume (%LGE) (r = 0.91, P < 
      .001). The 3D CNN-based quantification strongly correlated with manual 
      quantification of scar volume (r = 0.82-0.99, P < .001) and %LGE (r = 0.90-0.97, 
      P < .001) for all sites and vendors. The 3D CNN identified patients with a large 
      scar burden (>15%) with 98% accuracy (202 of 207) (95% confidence interval [CI]: 
      95%, 99%). When compared with 3D CNN, 2D CNN underestimated scar volume (r = 
      0.85, P < .001) and %LGE (r = 0.83, P < .001). The DSC of 3D CNN segmentation was 
      comparable among different vendors (P = .07) and higher than that of 2D CNN (DSC, 
      0.54 +/- 0.26 vs 0.48 +/- 0.29; P = .02). Conclusion In the hypertrophic 
      cardiomyopathy population, a three-dimensional convolutional neural network 
      enables fast and accurate quantification of myocardial scar volume, outperforms a 
      two-dimensional convolutional neural network, and demonstrates comparable 
      performance across different vendors. (c) RSNA, 2019 Online supplemental material 
      is available for this article.
FAU - Fahmy, Ahmed S
AU  - Fahmy AS
AUID- ORCID: 0000-0003-1992-6893
AD  - From the Departments of Medicine (Cardiovascular Division) (A.S.F., U.N., W.J.M., 
      R.N.) and Radiology (W.J.M.), Beth Israel Deaconess Medical Center and Harvard 
      Medical School, 330 Brookline Ave, Boston, MA 02215; Toronto General Hospital, 
      University Health Network, Toronto, Ontario, Canada (R.H.C.); and Hypertrophic 
      Cardiomyopathy Center, Division of Cardiology, Tufts Medical Center, Boston, Mass 
      (E.J.R., M.S.M.).
FAU - Neisius, Ulf
AU  - Neisius U
AUID- ORCID: 0000-0003-4744-9166
AD  - From the Departments of Medicine (Cardiovascular Division) (A.S.F., U.N., W.J.M., 
      R.N.) and Radiology (W.J.M.), Beth Israel Deaconess Medical Center and Harvard 
      Medical School, 330 Brookline Ave, Boston, MA 02215; Toronto General Hospital, 
      University Health Network, Toronto, Ontario, Canada (R.H.C.); and Hypertrophic 
      Cardiomyopathy Center, Division of Cardiology, Tufts Medical Center, Boston, Mass 
      (E.J.R., M.S.M.).
FAU - Chan, Raymond H
AU  - Chan RH
AD  - From the Departments of Medicine (Cardiovascular Division) (A.S.F., U.N., W.J.M., 
      R.N.) and Radiology (W.J.M.), Beth Israel Deaconess Medical Center and Harvard 
      Medical School, 330 Brookline Ave, Boston, MA 02215; Toronto General Hospital, 
      University Health Network, Toronto, Ontario, Canada (R.H.C.); and Hypertrophic 
      Cardiomyopathy Center, Division of Cardiology, Tufts Medical Center, Boston, Mass 
      (E.J.R., M.S.M.).
FAU - Rowin, Ethan J
AU  - Rowin EJ
AD  - From the Departments of Medicine (Cardiovascular Division) (A.S.F., U.N., W.J.M., 
      R.N.) and Radiology (W.J.M.), Beth Israel Deaconess Medical Center and Harvard 
      Medical School, 330 Brookline Ave, Boston, MA 02215; Toronto General Hospital, 
      University Health Network, Toronto, Ontario, Canada (R.H.C.); and Hypertrophic 
      Cardiomyopathy Center, Division of Cardiology, Tufts Medical Center, Boston, Mass 
      (E.J.R., M.S.M.).
FAU - Manning, Warren J
AU  - Manning WJ
AD  - From the Departments of Medicine (Cardiovascular Division) (A.S.F., U.N., W.J.M., 
      R.N.) and Radiology (W.J.M.), Beth Israel Deaconess Medical Center and Harvard 
      Medical School, 330 Brookline Ave, Boston, MA 02215; Toronto General Hospital, 
      University Health Network, Toronto, Ontario, Canada (R.H.C.); and Hypertrophic 
      Cardiomyopathy Center, Division of Cardiology, Tufts Medical Center, Boston, Mass 
      (E.J.R., M.S.M.).
FAU - Maron, Martin S
AU  - Maron MS
AD  - From the Departments of Medicine (Cardiovascular Division) (A.S.F., U.N., W.J.M., 
      R.N.) and Radiology (W.J.M.), Beth Israel Deaconess Medical Center and Harvard 
      Medical School, 330 Brookline Ave, Boston, MA 02215; Toronto General Hospital, 
      University Health Network, Toronto, Ontario, Canada (R.H.C.); and Hypertrophic 
      Cardiomyopathy Center, Division of Cardiology, Tufts Medical Center, Boston, Mass 
      (E.J.R., M.S.M.).
FAU - Nezafat, Reza
AU  - Nezafat R
AUID- ORCID: 0000-0002-1963-7138
AD  - From the Departments of Medicine (Cardiovascular Division) (A.S.F., U.N., W.J.M., 
      R.N.) and Radiology (W.J.M.), Beth Israel Deaconess Medical Center and Harvard 
      Medical School, 330 Brookline Ave, Boston, MA 02215; Toronto General Hospital, 
      University Health Network, Toronto, Ontario, Canada (R.H.C.); and Hypertrophic 
      Cardiomyopathy Center, Division of Cardiology, Tufts Medical Center, Boston, Mass 
      (E.J.R., M.S.M.).
LA  - eng
GR  - R01 HL129157/HL/NHLBI NIH HHS/United States
GR  - R01 HL129185/HL/NHLBI NIH HHS/United States
PT  - Journal Article
PT  - Multicenter Study
PT  - Research Support, N.I.H., Extramural
PT  - Research Support, Non-U.S. Gov't
DEP - 20191112
PL  - United States
TA  - Radiology
JT  - Radiology
JID - 0401260
SB  - IM
MH  - Adolescent
MH  - Adult
MH  - Aged
MH  - Aged, 80 and over
MH  - Cardiomyopathy, Hypertrophic/complications/*pathology
MH  - Child
MH  - Cicatrix/*diagnostic imaging/etiology
MH  - Female
MH  - Heart/diagnostic imaging
MH  - Humans
MH  - Image Interpretation, Computer-Assisted/*methods
MH  - Imaging, Three-Dimensional/*methods
MH  - Magnetic Resonance Imaging/*methods
MH  - Male
MH  - Middle Aged
MH  - Myocardium/pathology
MH  - *Neural Networks, Computer
MH  - Reproducibility of Results
MH  - Retrospective Studies
MH  - Young Adult
PMC - PMC6939743
EDAT- 2019/11/13 06:00
MHDA- 2020/04/21 06:00
PMCR- 2021/01/01
CRDT- 2019/11/13 06:00
PHST- 2019/11/13 06:00 [pubmed]
PHST- 2020/04/21 06:00 [medline]
PHST- 2019/11/13 06:00 [entrez]
PHST- 2021/01/01 00:00 [pmc-release]
AID - 10.1148/radiol.2019190737 [doi]
PST - ppublish
SO  - Radiology. 2020 Jan;294(1):52-60. doi: 10.1148/radiol.2019190737. Epub 2019 Nov 
      12.