Massive Dimensionality Reduction for the Left Ventricular Mesh

Lukasz Romaszko; Alan Lazarus; Hao Gao; Agnieszka Borowska; Xiaoyu Luo; Dirk Husmeier

doi:10.11159/icsta19.24

Massive Dimensionality Reduction for the Left Ventricular Mesh

Lukasz Romaszko, Alan Lazarus, Hao Gao, Agnieszka Borowska, Xiaoyu Luo, Dirk Husmeier

Econometrics and Data Science

Research output: Chapter in Book / Report / Conference proceeding › Conference contribution › Academic › peer-review

7 Downloads (Pure)

Abstract

Statistical emulation is a promising approach for the translation of cardio-mechanical modelling into the clinical practice. However, a key challenge is to find a low-dimensional representation of the heart, or, for the specific purpose of diagnosing the risk of heart attacks, the left-ventricle of the heart. We consider the problem of dimensionality reduction of the left ventricular mesh, in which we investigate three classes of techniques: principal component analysis (PCA), deep learning (DL) methods based on auto-encoders, and a parametric model from the cardio-mechanical literature. Our finding is that PCA performs as well as the computationally more expensive DL methods, and both outperform the state-of-the-art parametric model.

Original language	English
Title of host publication	Proceedings of the International Conference on Statistics: Theory and Applications (ICSTA'19)
Editors	Gangaram S. Ladde, Daniel Jeske
Publisher	Avestia Publishing
Pages	1-8
Number of pages	8
ISBN (Print)	9781927877647
DOIs	https://doi.org/10.11159/icsta19.24
Publication status	Published - 2019
Event	International Conference of Statistics: Theory and Applications, ICSTA 2019 - Lisbon, Portugal Duration: 13 Aug 2019 → 14 Aug 2019

Publication series

Name	Proceedings of the International Conference on Statistics
ISSN (Electronic)	2562-7767

Conference

Conference	International Conference of Statistics: Theory and Applications, ICSTA 2019
Country/Territory	Portugal
City	Lisbon
Period	13/08/19 → 14/08/19

Bibliographical note

Funding Information:
This work was funded by the UK Engineering and Physical Sciences Research Council (EPSRC), grant number EP/N014642/1. Alan Lazarus is partially funded by a grant from GlaxoSmithKline plc. Dirk Husmeier is supported by a grant from the Royal Society of Edinburgh, award number 62335.

Publisher Copyright:
© 2019, Avestia Publishing. All rights reserved.

Keywords

auto-encoders
deep learning
Dimensionality reduction
left ventricular mesh
principal component analysis

Access to Document

10.11159/icsta19.24

ICSTA-24_Massive Dimensionality Reduction for the Left Ventricular MeshFinal published version, 2.09 MBLicence: Article 25fa Dutch Copyright Act

Persistent URL (handle)

Persistent link

Cite this

Romaszko, L., Lazarus, A., Gao, H., Borowska, A., Luo, X., & Husmeier, D. (2019). Massive Dimensionality Reduction for the Left Ventricular Mesh. In G. S. Ladde, & D. Jeske (Eds.), Proceedings of the International Conference on Statistics: Theory and Applications (ICSTA'19) (pp. 1-8). Article 24 (Proceedings of the International Conference on Statistics). Avestia Publishing. https://doi.org/10.11159/icsta19.24

@inproceedings{bd8bc8c01c5149ec96093a1b434d3ea3,

title = "Massive Dimensionality Reduction for the Left Ventricular Mesh",

abstract = "Statistical emulation is a promising approach for the translation of cardio-mechanical modelling into the clinical practice. However, a key challenge is to find a low-dimensional representation of the heart, or, for the specific purpose of diagnosing the risk of heart attacks, the left-ventricle of the heart. We consider the problem of dimensionality reduction of the left ventricular mesh, in which we investigate three classes of techniques: principal component analysis (PCA), deep learning (DL) methods based on auto-encoders, and a parametric model from the cardio-mechanical literature. Our finding is that PCA performs as well as the computationally more expensive DL methods, and both outperform the state-of-the-art parametric model.",

keywords = "auto-encoders, deep learning, Dimensionality reduction, left ventricular mesh, principal component analysis",

author = "Lukasz Romaszko and Alan Lazarus and Hao Gao and Agnieszka Borowska and Xiaoyu Luo and Dirk Husmeier",

note = "Funding Information: This work was funded by the UK Engineering and Physical Sciences Research Council (EPSRC), grant number EP/N014642/1. Alan Lazarus is partially funded by a grant from GlaxoSmithKline plc. Dirk Husmeier is supported by a grant from the Royal Society of Edinburgh, award number 62335. Publisher Copyright: {\textcopyright} 2019, Avestia Publishing. All rights reserved.; International Conference of Statistics: Theory and Applications, ICSTA 2019 ; Conference date: 13-08-2019 Through 14-08-2019",

year = "2019",

doi = "10.11159/icsta19.24",

language = "English",

isbn = "9781927877647",

series = "Proceedings of the International Conference on Statistics",

publisher = "Avestia Publishing",

pages = "1--8",

editor = "Ladde, {Gangaram S.} and Daniel Jeske",

booktitle = "Proceedings of the International Conference on Statistics: Theory and Applications (ICSTA'19)",

}

Romaszko, L, Lazarus, A, Gao, H, Borowska, A, Luo, X & Husmeier, D 2019, Massive Dimensionality Reduction for the Left Ventricular Mesh. in GS Ladde & D Jeske (eds), Proceedings of the International Conference on Statistics: Theory and Applications (ICSTA'19)., 24, Proceedings of the International Conference on Statistics, Avestia Publishing, pp. 1-8, International Conference of Statistics: Theory and Applications, ICSTA 2019, Lisbon, Portugal, 13/08/19. https://doi.org/10.11159/icsta19.24

Massive Dimensionality Reduction for the Left Ventricular Mesh. / Romaszko, Lukasz; Lazarus, Alan; Gao, Hao et al.
Proceedings of the International Conference on Statistics: Theory and Applications (ICSTA'19). ed. / Gangaram S. Ladde; Daniel Jeske. Avestia Publishing, 2019. p. 1-8 24 (Proceedings of the International Conference on Statistics).

Research output: Chapter in Book / Report / Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Massive Dimensionality Reduction for the Left Ventricular Mesh

AU - Romaszko, Lukasz

AU - Lazarus, Alan

AU - Gao, Hao

AU - Borowska, Agnieszka

AU - Luo, Xiaoyu

AU - Husmeier, Dirk

N1 - Funding Information: This work was funded by the UK Engineering and Physical Sciences Research Council (EPSRC), grant number EP/N014642/1. Alan Lazarus is partially funded by a grant from GlaxoSmithKline plc. Dirk Husmeier is supported by a grant from the Royal Society of Edinburgh, award number 62335. Publisher Copyright: © 2019, Avestia Publishing. All rights reserved.

PY - 2019

Y1 - 2019

N2 - Statistical emulation is a promising approach for the translation of cardio-mechanical modelling into the clinical practice. However, a key challenge is to find a low-dimensional representation of the heart, or, for the specific purpose of diagnosing the risk of heart attacks, the left-ventricle of the heart. We consider the problem of dimensionality reduction of the left ventricular mesh, in which we investigate three classes of techniques: principal component analysis (PCA), deep learning (DL) methods based on auto-encoders, and a parametric model from the cardio-mechanical literature. Our finding is that PCA performs as well as the computationally more expensive DL methods, and both outperform the state-of-the-art parametric model.

AB - Statistical emulation is a promising approach for the translation of cardio-mechanical modelling into the clinical practice. However, a key challenge is to find a low-dimensional representation of the heart, or, for the specific purpose of diagnosing the risk of heart attacks, the left-ventricle of the heart. We consider the problem of dimensionality reduction of the left ventricular mesh, in which we investigate three classes of techniques: principal component analysis (PCA), deep learning (DL) methods based on auto-encoders, and a parametric model from the cardio-mechanical literature. Our finding is that PCA performs as well as the computationally more expensive DL methods, and both outperform the state-of-the-art parametric model.

KW - auto-encoders

KW - deep learning

KW - Dimensionality reduction

KW - left ventricular mesh

KW - principal component analysis

UR - http://www.scopus.com/inward/record.url?scp=85113480175&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85113480175&partnerID=8YFLogxK

UR - https://avestia.com/ICSTA2019_Proceedings/

U2 - 10.11159/icsta19.24

DO - 10.11159/icsta19.24

M3 - Conference contribution

AN - SCOPUS:85113480175

SN - 9781927877647

T3 - Proceedings of the International Conference on Statistics

SP - 1

EP - 8

BT - Proceedings of the International Conference on Statistics: Theory and Applications (ICSTA'19)

A2 - Ladde, Gangaram S.

A2 - Jeske, Daniel

PB - Avestia Publishing

T2 - International Conference of Statistics: Theory and Applications, ICSTA 2019

Y2 - 13 August 2019 through 14 August 2019

ER -

Romaszko L, Lazarus A, Gao H, Borowska A, Luo X, Husmeier D. Massive Dimensionality Reduction for the Left Ventricular Mesh. In Ladde GS, Jeske D, editors, Proceedings of the International Conference on Statistics: Theory and Applications (ICSTA'19). Avestia Publishing. 2019. p. 1-8. 24. (Proceedings of the International Conference on Statistics). Epub 2019 Aug 13. doi: 10.11159/icsta19.24

Massive Dimensionality Reduction for the Left Ventricular Mesh

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Persistent URL (handle)

Other files and links

Fingerprint

Cite this