학술논문
Interpretable machine learning for automated left ventricular scar quantification in hypertrophic cardiomyopathy patients.
Document Type
Academic Journal
Author
Navidi Z; Division of Cardiology, Peter Munk Cardiac Center, Toronto General Hospital, University Health Network, University of Toronto, Toronto, Canada.; Department of Computer Science, University of Toronto, Toronto, Canada.; Vector Institute, Toronto, Canada.; Sun J; Division of Cardiology, Peter Munk Cardiac Center, Toronto General Hospital, University Health Network, University of Toronto, Toronto, Canada.; Chan RH; Division of Cardiology, Peter Munk Cardiac Center, Toronto General Hospital, University Health Network, University of Toronto, Toronto, Canada.; Hanneman K; Department of Radiology, University Health Network, University of Toronto, Toronto, Canada.; Al-Arnawoot A; Department of Radiology, University Health Network, University of Toronto, Toronto, Canada.; Munim A; Vector Institute, Toronto, Canada.; Rakowski H; Division of Cardiology, Peter Munk Cardiac Center, Toronto General Hospital, University Health Network, University of Toronto, Toronto, Canada.; Maron MS; Division of Cardiology, Tufts Medical Center, Boston, United States of America.; Woo A; Division of Cardiology, Peter Munk Cardiac Center, Toronto General Hospital, University Health Network, University of Toronto, Toronto, Canada.; Wang B; Division of Cardiology, Peter Munk Cardiac Center, Toronto General Hospital, University Health Network, University of Toronto, Toronto, Canada.; Department of Computer Science, University of Toronto, Toronto, Canada.; Vector Institute, Toronto, Canada.; Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Canada.; Tsang W; Division of Cardiology, Peter Munk Cardiac Center, Toronto General Hospital, University Health Network, University of Toronto, Toronto, Canada.
Source
Publisher: PLOS Country of Publication: United States NLM ID: 9918335064206676 Publication Model: eCollection Cited Medium: Internet ISSN: 2767-3170 (Electronic) Linking ISSN: 27673170 NLM ISO Abbreviation: PLOS Digit Health Subsets: PubMed not MEDLINE
Subject
Language
English
Abstract
Scar quantification on cardiovascular magnetic resonance (CMR) late gadolinium enhancement (LGE) images is important in risk stratifying patients with hypertrophic cardiomyopathy (HCM) due to the importance of scar burden in predicting clinical outcomes. We aimed to develop a machine learning (ML) model that contours left ventricular (LV) endo- and epicardial borders and quantifies CMR LGE images from HCM patients.We retrospectively studied 2557 unprocessed images from 307 HCM patients followed at the University Health Network (Canada) and Tufts Medical Center (USA). LGE images were manually segmented by two experts using two different software packages. Using 6SD LGE intensity cutoff as the gold standard, a 2-dimensional convolutional neural network (CNN) was trained on 80% and tested on the remaining 20% of the data. Model performance was evaluated using the Dice Similarity Coefficient (DSC), Bland-Altman, and Pearson's correlation. The 6SD model DSC scores were good to excellent at 0.91 ± 0.04, 0.83 ± 0.03, and 0.64 ± 0.09 for the LV endocardium, epicardium, and scar segmentation, respectively. The bias and limits of agreement for the percentage of LGE to LV mass were low (-0.53 ± 2.71%), and correlation high (r = 0.92). This fully automated interpretable ML algorithm allows rapid and accurate scar quantification from CMR LGE images. This program does not require manual image pre-processing, and was trained with multiple experts and software, increasing its generalizability.
Competing Interests: The authors declare that they have no competing interests.
(Copyright: © 2023 Navidi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)
Competing Interests: The authors declare that they have no competing interests.
(Copyright: © 2023 Navidi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)