Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

MVnet : automated time-resolved tracking of the mitral valve plane in CMR long-axis cine images with residual neural networks: a multi-center, multi-vendor study

Gonzales, Ricardo A. LU orcid ; Seemann, Felicia LU ; Lamy, Jérôme ; Mojibian, Hamid ; Atar, Dan ; Erlinge, David LU orcid ; Steding-Ehrenborg, Katarina LU ; Arheden, Håkan LU ; Hu, Chenxi and Onofrey, John A. , et al. (2021) In Journal of Cardiovascular Magnetic Resonance 23. p.1-15
Abstract

Background: Mitral annular plane systolic excursion (MAPSE) and left ventricular (LV) early diastolic velocity (e’) are key metrics of systolic and diastolic function, but not often measured by cardiovascular magnetic resonance (CMR). Its derivation is possible with manual, precise annotation of the mitral valve (MV) insertion points along the cardiac cycle in both two and four-chamber long-axis cines, but this process is highly time-consuming, laborious, and prone to errors. A fully automated, consistent, fast, and accurate method for MV plane tracking is lacking. In this study, we propose MVnet, a deep learning approach for MV point localization and tracking capable of deriving such clinical metrics comparable to human expert-level... (More)

Background: Mitral annular plane systolic excursion (MAPSE) and left ventricular (LV) early diastolic velocity (e’) are key metrics of systolic and diastolic function, but not often measured by cardiovascular magnetic resonance (CMR). Its derivation is possible with manual, precise annotation of the mitral valve (MV) insertion points along the cardiac cycle in both two and four-chamber long-axis cines, but this process is highly time-consuming, laborious, and prone to errors. A fully automated, consistent, fast, and accurate method for MV plane tracking is lacking. In this study, we propose MVnet, a deep learning approach for MV point localization and tracking capable of deriving such clinical metrics comparable to human expert-level performance, and validated it in a multi-vendor, multi-center clinical population. Methods: The proposed pipeline first performs a coarse MV point annotation in a given cine accurately enough to apply an automated linear transformation task, which standardizes the size, cropping, resolution, and heart orientation, and second, tracks the MV points with high accuracy. The model was trained and evaluated on 38,854 cine images from 703 patients with diverse cardiovascular conditions, scanned on equipment from 3 main vendors, 16 centers, and 7 countries, and manually annotated by 10 observers. Agreement was assessed by the intra-class correlation coefficient (ICC) for both clinical metrics and by the distance error in the MV plane displacement. For inter-observer variability analysis, an additional pair of observers performed manual annotations in a randomly chosen set of 50 patients. Results: MVnet achieved a fast segmentation (<1 s/cine) with excellent ICCs of 0.94 (MAPSE) and 0.93 (LV e’) and a MV plane tracking error of −0.10 ± 0.97 mm. In a similar manner, the inter-observer variability analysis yielded ICCs of 0.95 and 0.89 and a tracking error of −0.15 ± 1.18 mm, respectively. Conclusion: A dual-stage deep learning approach for automated annotation of MV points for systolic and diastolic evaluation in CMR long-axis cine images was developed. The method is able to carefully track these points with high accuracy and in a timely manner. This will improve the feasibility of CMR methods which rely on valve tracking and increase their utility in a clinical setting.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; ; and , et al. (More)
; ; ; ; ; ; ; ; ; ; and (Less)
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Annotation, Left ventricular dysfunction, Residual neural networks
in
Journal of Cardiovascular Magnetic Resonance
volume
23
article number
137
pages
1 - 15
publisher
BioMed Central (BMC)
external identifiers
  • pmid:34857009
  • scopus:85120752806
ISSN
1097-6647
DOI
10.1186/s12968-021-00824-2
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2021, The Author(s).
id
3ac53ce5-5f4e-4815-8c37-252b6bb7ec2f
date added to LUP
2022-10-21 10:18:11
date last changed
2024-04-18 15:08:16
@article{3ac53ce5-5f4e-4815-8c37-252b6bb7ec2f,
  abstract     = {{<p>Background: Mitral annular plane systolic excursion (MAPSE) and left ventricular (LV) early diastolic velocity (e’) are key metrics of systolic and diastolic function, but not often measured by cardiovascular magnetic resonance (CMR). Its derivation is possible with manual, precise annotation of the mitral valve (MV) insertion points along the cardiac cycle in both two and four-chamber long-axis cines, but this process is highly time-consuming, laborious, and prone to errors. A fully automated, consistent, fast, and accurate method for MV plane tracking is lacking. In this study, we propose MVnet, a deep learning approach for MV point localization and tracking capable of deriving such clinical metrics comparable to human expert-level performance, and validated it in a multi-vendor, multi-center clinical population. Methods: The proposed pipeline first performs a coarse MV point annotation in a given cine accurately enough to apply an automated linear transformation task, which standardizes the size, cropping, resolution, and heart orientation, and second, tracks the MV points with high accuracy. The model was trained and evaluated on 38,854 cine images from 703 patients with diverse cardiovascular conditions, scanned on equipment from 3 main vendors, 16 centers, and 7 countries, and manually annotated by 10 observers. Agreement was assessed by the intra-class correlation coefficient (ICC) for both clinical metrics and by the distance error in the MV plane displacement. For inter-observer variability analysis, an additional pair of observers performed manual annotations in a randomly chosen set of 50 patients. Results: MVnet achieved a fast segmentation (&lt;1 s/cine) with excellent ICCs of 0.94 (MAPSE) and 0.93 (LV e’) and a MV plane tracking error of −0.10 ± 0.97 mm. In a similar manner, the inter-observer variability analysis yielded ICCs of 0.95 and 0.89 and a tracking error of −0.15 ± 1.18 mm, respectively. Conclusion: A dual-stage deep learning approach for automated annotation of MV points for systolic and diastolic evaluation in CMR long-axis cine images was developed. The method is able to carefully track these points with high accuracy and in a timely manner. This will improve the feasibility of CMR methods which rely on valve tracking and increase their utility in a clinical setting.</p>}},
  author       = {{Gonzales, Ricardo A. and Seemann, Felicia and Lamy, Jérôme and Mojibian, Hamid and Atar, Dan and Erlinge, David and Steding-Ehrenborg, Katarina and Arheden, Håkan and Hu, Chenxi and Onofrey, John A. and Peters, Dana C. and Heiberg, Einar}},
  issn         = {{1097-6647}},
  keywords     = {{Annotation; Left ventricular dysfunction; Residual neural networks}},
  language     = {{eng}},
  pages        = {{1--15}},
  publisher    = {{BioMed Central (BMC)}},
  series       = {{Journal of Cardiovascular Magnetic Resonance}},
  title        = {{MVnet : automated time-resolved tracking of the mitral valve plane in CMR long-axis cine images with residual neural networks: a multi-center, multi-vendor study}},
  url          = {{http://dx.doi.org/10.1186/s12968-021-00824-2}},
  doi          = {{10.1186/s12968-021-00824-2}},
  volume       = {{23}},
  year         = {{2021}},
}