King's College London

Research portal

An Implementation of Patient-Specific Biventricular Mechanics Simulations With a Deep Learning and Computational Pipeline

Research output: Contribution to journalArticlepeer-review

Standard

An Implementation of Patient-Specific Biventricular Mechanics Simulations With a Deep Learning and Computational Pipeline. / Miller, Renee; Kerfoot, Eric; Mauger, Charlène et al.

In: Frontiers in Physiology, Vol. 12, 716597, 16.09.2021.

Research output: Contribution to journalArticlepeer-review

Harvard

Miller, R, Kerfoot, E, Mauger, C, Ismail, T, Young, A & Nordsletten, D 2021, 'An Implementation of Patient-Specific Biventricular Mechanics Simulations With a Deep Learning and Computational Pipeline', Frontiers in Physiology, vol. 12, 716597. https://doi.org/10.3389/fphys.2021.716597

APA

Miller, R., Kerfoot, E., Mauger, C., Ismail, T., Young, A., & Nordsletten, D. (2021). An Implementation of Patient-Specific Biventricular Mechanics Simulations With a Deep Learning and Computational Pipeline. Frontiers in Physiology, 12, [716597]. https://doi.org/10.3389/fphys.2021.716597

Vancouver

Miller R, Kerfoot E, Mauger C, Ismail T, Young A, Nordsletten D. An Implementation of Patient-Specific Biventricular Mechanics Simulations With a Deep Learning and Computational Pipeline. Frontiers in Physiology. 2021 Sep 16;12. 716597. https://doi.org/10.3389/fphys.2021.716597

Author

Miller, Renee ; Kerfoot, Eric ; Mauger, Charlène et al. / An Implementation of Patient-Specific Biventricular Mechanics Simulations With a Deep Learning and Computational Pipeline. In: Frontiers in Physiology. 2021 ; Vol. 12.

Bibtex Download

@article{43aa6298bb1e4049b2f828ac2168cd26,
title = "An Implementation of Patient-Specific Biventricular Mechanics Simulations With a Deep Learning and Computational Pipeline",
abstract = "Parameterised patient-specific models of the heart enable quantitative analysis of cardiac function as well as estimation of regional stress and intrinsic tissue stiffness. However, the development of personalised models and subsequent simulations have often required lengthy manual setup, from image labelling through to generating the finite element model and assigning boundary conditions. Recently, rapid patient-specific finite element modelling has been made possible through the use of machine learning techniques. In this paper, utilising multiple neural networks for image labelling and detection of valve landmarks, together with streamlined data integration, a pipeline for generating patient-specific biventricular models is applied to clinically-acquired data from a diverse cohort of individuals, including hypertrophic and dilated cardiomyopathy patients and healthy volunteers. Valve motion from tracked landmarks as well as cavity volumes measured from labelled images are used to drive realistic motion and estimate passive tissue stiffness values. The neural networks are shown to accurately label cardiac regions and features for these diverse morphologies. Furthermore, differences in global intrinsic parameters, such as tissue anisotropy and normalised active tension, between groups illustrate respective underlying changes in tissue composition and/or structure as a result of pathology. This study shows the successful application of a generic pipeline for biventricular modelling, incorporating artificial intelligence solutions, within a diverse cohort.",
author = "Renee Miller and Eric Kerfoot and Charl{\`e}ne Mauger and Tevfik Ismail and Alistair Young and David Nordsletten",
note = "Funding Information: DN would like to acknowledge funding from Engineering and Physical Sciences Research Council (EP/N011554/1 and EP/R003866/1). This work was also supported by the Wellcome ESPRC Centre for Medical Engineering at King{\textquoteright}s College London (WT 203148/Z/16/Z) and the British Heart Foundation (TG/17/3/33406). AY would like to acknowledge funding from the National Heart, Lung and Blood Institute (NIH R01HL121754). Publisher Copyright: {\textcopyright} Copyright {\textcopyright} 2021 Miller, Kerfoot, Mauger, Ismail, Young and Nordsletten.",
year = "2021",
month = sep,
day = "16",
doi = "10.3389/fphys.2021.716597",
language = "English",
volume = "12",
journal = "Frontiers in Physiology",
issn = "1664-042X",
publisher = "Frontiers",

}

RIS (suitable for import to EndNote) Download

TY - JOUR

T1 - An Implementation of Patient-Specific Biventricular Mechanics Simulations With a Deep Learning and Computational Pipeline

AU - Miller, Renee

AU - Kerfoot, Eric

AU - Mauger, Charlène

AU - Ismail, Tevfik

AU - Young, Alistair

AU - Nordsletten, David

N1 - Funding Information: DN would like to acknowledge funding from Engineering and Physical Sciences Research Council (EP/N011554/1 and EP/R003866/1). This work was also supported by the Wellcome ESPRC Centre for Medical Engineering at King’s College London (WT 203148/Z/16/Z) and the British Heart Foundation (TG/17/3/33406). AY would like to acknowledge funding from the National Heart, Lung and Blood Institute (NIH R01HL121754). Publisher Copyright: © Copyright © 2021 Miller, Kerfoot, Mauger, Ismail, Young and Nordsletten.

PY - 2021/9/16

Y1 - 2021/9/16

N2 - Parameterised patient-specific models of the heart enable quantitative analysis of cardiac function as well as estimation of regional stress and intrinsic tissue stiffness. However, the development of personalised models and subsequent simulations have often required lengthy manual setup, from image labelling through to generating the finite element model and assigning boundary conditions. Recently, rapid patient-specific finite element modelling has been made possible through the use of machine learning techniques. In this paper, utilising multiple neural networks for image labelling and detection of valve landmarks, together with streamlined data integration, a pipeline for generating patient-specific biventricular models is applied to clinically-acquired data from a diverse cohort of individuals, including hypertrophic and dilated cardiomyopathy patients and healthy volunteers. Valve motion from tracked landmarks as well as cavity volumes measured from labelled images are used to drive realistic motion and estimate passive tissue stiffness values. The neural networks are shown to accurately label cardiac regions and features for these diverse morphologies. Furthermore, differences in global intrinsic parameters, such as tissue anisotropy and normalised active tension, between groups illustrate respective underlying changes in tissue composition and/or structure as a result of pathology. This study shows the successful application of a generic pipeline for biventricular modelling, incorporating artificial intelligence solutions, within a diverse cohort.

AB - Parameterised patient-specific models of the heart enable quantitative analysis of cardiac function as well as estimation of regional stress and intrinsic tissue stiffness. However, the development of personalised models and subsequent simulations have often required lengthy manual setup, from image labelling through to generating the finite element model and assigning boundary conditions. Recently, rapid patient-specific finite element modelling has been made possible through the use of machine learning techniques. In this paper, utilising multiple neural networks for image labelling and detection of valve landmarks, together with streamlined data integration, a pipeline for generating patient-specific biventricular models is applied to clinically-acquired data from a diverse cohort of individuals, including hypertrophic and dilated cardiomyopathy patients and healthy volunteers. Valve motion from tracked landmarks as well as cavity volumes measured from labelled images are used to drive realistic motion and estimate passive tissue stiffness values. The neural networks are shown to accurately label cardiac regions and features for these diverse morphologies. Furthermore, differences in global intrinsic parameters, such as tissue anisotropy and normalised active tension, between groups illustrate respective underlying changes in tissue composition and/or structure as a result of pathology. This study shows the successful application of a generic pipeline for biventricular modelling, incorporating artificial intelligence solutions, within a diverse cohort.

UR - http://www.scopus.com/inward/record.url?scp=85116305875&partnerID=8YFLogxK

U2 - 10.3389/fphys.2021.716597

DO - 10.3389/fphys.2021.716597

M3 - Article

VL - 12

JO - Frontiers in Physiology

JF - Frontiers in Physiology

SN - 1664-042X

M1 - 716597

ER -

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454