No Cover Image

Journal article 63 views 6 downloads

Artificial intelligence approaches to predict coronary stenosis severity using non-invasive fractional flow reserve / Jason M. Carson; Neeraj Kavan Chakshu; Igor Sazonov; Perumal Nithiarasu

Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine, Volume: 234, Issue: 11, Pages: 1337 - 1350

Swansea University Authors: Jason M., Carson, Neeraj Kavan, Chakshu, Igor, Sazonov, Perumal, Nithiarasu

  • 0954411920946526.pdf

    PDF | Version of Record

    Distributed under the terms of a Creative Commons Attribution Non Commercial (CC-BY-NC) License.

    Download (1.05MB)

Abstract

Fractional flow reserve is the current reference standard in the assessment of the functional impact of a stenosis in coronary heart disease. In this study, three models of artificial intelligence of varying degrees of complexity were compared to fractional flow reserve measurements. The three model...

Full description

Published in: Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine
ISSN: 0954-4119 2041-3033
Published: SAGE Publications 2020
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa54883
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2020-08-05T12:29:26Z
last_indexed 2020-11-20T04:10:49Z
id cronfa54883
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2020-11-19T14:46:03.4141832</datestamp><bib-version>v2</bib-version><id>54883</id><entry>2020-08-05</entry><title>Artificial intelligence approaches to predict coronary stenosis severity using non-invasive fractional flow reserve</title><swanseaauthors><author><sid>d0fe636d559f9023182e4315c2940595</sid><ORCID>0000-0001-6634-9123</ORCID><firstname>Jason M.</firstname><surname>Carson</surname><name>Jason M. Carson</name><active>true</active><ethesisStudent>true</ethesisStudent></author><author><sid>e21c85ee9062e9be0fff8ab9d77b14d7</sid><firstname>Neeraj Kavan</firstname><surname>Chakshu</surname><name>Neeraj Kavan Chakshu</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>05a507952e26462561085fb6f62c8897</sid><ORCID>0000-0001-6685-2351</ORCID><firstname>Igor</firstname><surname>Sazonov</surname><name>Igor Sazonov</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>3b28bf59358fc2b9bd9a46897dbfc92d</sid><ORCID>0000-0002-4901-2980</ORCID><firstname>Perumal</firstname><surname>Nithiarasu</surname><name>Perumal Nithiarasu</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2020-08-05</date><abstract>Fractional flow reserve is the current reference standard in the assessment of the functional impact of a stenosis in coronary heart disease. In this study, three models of artificial intelligence of varying degrees of complexity were compared to fractional flow reserve measurements. The three models are the multivariate polynomial regression, which is a statistical method used primarily for correlation; the feed-forward neural network; and the long short-term memory, which is a type of recurrent neural network that is suited to modelling sequences. The models were initially trained using a virtual patient database that was generated from a validated one-dimensional physics-based model. The feed-forward neural network performed the best for all test cases considered, which were a single vessel case from a virtual patient database, a multi-vessel network from a virtual patient database, and 25 clinically invasive fractional flow reserve measurements from real patients. The feed-forward neural network model achieved around 99% diagnostic accuracy in both tests involving virtual patients, and a respectable 72% diagnostic accuracy when compared to the invasive fractional flow reserve measurements. The multivariate polynomial regression model performed well in the single vessel case, but struggled on network cases as the variation of input features was much larger. The long short-term memory performed well for the single vessel cases, but tended to have a bias towards a positive fractional flow reserve prediction for the virtual multi-vessel case, and for the patient cases. Overall, the feed-forward neural network shows promise in successfully predicting fractional flow reserve in real patients, and could be a viable option if trained using a large enough data set of real patients.</abstract><type>Journal Article</type><journal>Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine</journal><volume>234</volume><journalNumber>11</journalNumber><paginationStart>1337</paginationStart><paginationEnd>1350</paginationEnd><publisher>SAGE Publications</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0954-4119</issnPrint><issnElectronic>2041-3033</issnElectronic><keywords>Artificial intelligence, computational mechanics, biomedical engineering, haemodynamic modelling, coronary heart disease, fractional flow reserve</keywords><publishedDay>1</publishedDay><publishedMonth>11</publishedMonth><publishedYear>2020</publishedYear><publishedDate>2020-11-01</publishedDate><doi>10.1177/0954411920946526</doi><url/><notes/><college>COLLEGE NANME</college><department>College of Engineering</department><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><lastEdited>2020-11-19T14:46:03.4141832</lastEdited><Created>2020-08-05T13:24:12.4388500</Created><path><level id="1">College of Engineering</level><level id="2">Engineering</level></path><authors><author><firstname>Jason M.</firstname><surname>Carson</surname><orcid>0000-0001-6634-9123</orcid><order>1</order></author><author><firstname>Neeraj Kavan</firstname><surname>Chakshu</surname><order>2</order></author><author><firstname>Igor</firstname><surname>Sazonov</surname><orcid>0000-0001-6685-2351</orcid><order>3</order></author><author><firstname>Perumal</firstname><surname>Nithiarasu</surname><orcid>0000-0002-4901-2980</orcid><order>4</order></author></authors><documents><document><filename>54883__18453__1e0a5433b399441b8fc2a59dde428cbf.pdf</filename><originalFilename>0954411920946526.pdf</originalFilename><uploaded>2020-10-19T12:58:39.6087533</uploaded><type>Output</type><contentLength>1099181</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><action/><documentNotes>Distributed under the terms of a Creative Commons Attribution Non Commercial (CC-BY-NC) License.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language></document></documents><OutputDurs/></rfc1807>
spelling 2020-11-19T14:46:03.4141832 v2 54883 2020-08-05 Artificial intelligence approaches to predict coronary stenosis severity using non-invasive fractional flow reserve d0fe636d559f9023182e4315c2940595 0000-0001-6634-9123 Jason M. Carson Jason M. Carson true true e21c85ee9062e9be0fff8ab9d77b14d7 Neeraj Kavan Chakshu Neeraj Kavan Chakshu true false 05a507952e26462561085fb6f62c8897 0000-0001-6685-2351 Igor Sazonov Igor Sazonov true false 3b28bf59358fc2b9bd9a46897dbfc92d 0000-0002-4901-2980 Perumal Nithiarasu Perumal Nithiarasu true false 2020-08-05 Fractional flow reserve is the current reference standard in the assessment of the functional impact of a stenosis in coronary heart disease. In this study, three models of artificial intelligence of varying degrees of complexity were compared to fractional flow reserve measurements. The three models are the multivariate polynomial regression, which is a statistical method used primarily for correlation; the feed-forward neural network; and the long short-term memory, which is a type of recurrent neural network that is suited to modelling sequences. The models were initially trained using a virtual patient database that was generated from a validated one-dimensional physics-based model. The feed-forward neural network performed the best for all test cases considered, which were a single vessel case from a virtual patient database, a multi-vessel network from a virtual patient database, and 25 clinically invasive fractional flow reserve measurements from real patients. The feed-forward neural network model achieved around 99% diagnostic accuracy in both tests involving virtual patients, and a respectable 72% diagnostic accuracy when compared to the invasive fractional flow reserve measurements. The multivariate polynomial regression model performed well in the single vessel case, but struggled on network cases as the variation of input features was much larger. The long short-term memory performed well for the single vessel cases, but tended to have a bias towards a positive fractional flow reserve prediction for the virtual multi-vessel case, and for the patient cases. Overall, the feed-forward neural network shows promise in successfully predicting fractional flow reserve in real patients, and could be a viable option if trained using a large enough data set of real patients. Journal Article Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine 234 11 1337 1350 SAGE Publications 0954-4119 2041-3033 Artificial intelligence, computational mechanics, biomedical engineering, haemodynamic modelling, coronary heart disease, fractional flow reserve 1 11 2020 2020-11-01 10.1177/0954411920946526 COLLEGE NANME College of Engineering COLLEGE CODE Swansea University 2020-11-19T14:46:03.4141832 2020-08-05T13:24:12.4388500 College of Engineering Engineering Jason M. Carson 0000-0001-6634-9123 1 Neeraj Kavan Chakshu 2 Igor Sazonov 0000-0001-6685-2351 3 Perumal Nithiarasu 0000-0002-4901-2980 4 54883__18453__1e0a5433b399441b8fc2a59dde428cbf.pdf 0954411920946526.pdf 2020-10-19T12:58:39.6087533 Output 1099181 application/pdf Version of Record true Distributed under the terms of a Creative Commons Attribution Non Commercial (CC-BY-NC) License. true eng
title Artificial intelligence approaches to predict coronary stenosis severity using non-invasive fractional flow reserve
spellingShingle Artificial intelligence approaches to predict coronary stenosis severity using non-invasive fractional flow reserve
Jason M., Carson
Neeraj Kavan, Chakshu
Igor, Sazonov
Perumal, Nithiarasu
title_short Artificial intelligence approaches to predict coronary stenosis severity using non-invasive fractional flow reserve
title_full Artificial intelligence approaches to predict coronary stenosis severity using non-invasive fractional flow reserve
title_fullStr Artificial intelligence approaches to predict coronary stenosis severity using non-invasive fractional flow reserve
title_full_unstemmed Artificial intelligence approaches to predict coronary stenosis severity using non-invasive fractional flow reserve
title_sort Artificial intelligence approaches to predict coronary stenosis severity using non-invasive fractional flow reserve
author_id_str_mv d0fe636d559f9023182e4315c2940595
e21c85ee9062e9be0fff8ab9d77b14d7
05a507952e26462561085fb6f62c8897
3b28bf59358fc2b9bd9a46897dbfc92d
author_id_fullname_str_mv d0fe636d559f9023182e4315c2940595_***_Jason M., Carson
e21c85ee9062e9be0fff8ab9d77b14d7_***_Neeraj Kavan, Chakshu
05a507952e26462561085fb6f62c8897_***_Igor, Sazonov
3b28bf59358fc2b9bd9a46897dbfc92d_***_Perumal, Nithiarasu
author Jason M., Carson
Neeraj Kavan, Chakshu
Igor, Sazonov
Perumal, Nithiarasu
author2 Jason M. Carson
Neeraj Kavan Chakshu
Igor Sazonov
Perumal Nithiarasu
format Journal article
container_title Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine
container_volume 234
container_issue 11
container_start_page 1337
publishDate 2020
institution Swansea University
issn 0954-4119
2041-3033
doi_str_mv 10.1177/0954411920946526
publisher SAGE Publications
college_str College of Engineering
hierarchytype
hierarchy_top_id collegeofengineering
hierarchy_top_title College of Engineering
hierarchy_parent_id collegeofengineering
hierarchy_parent_title College of Engineering
department_str Engineering{{{_:::_}}}College of Engineering{{{_:::_}}}Engineering
document_store_str 1
active_str 0
description Fractional flow reserve is the current reference standard in the assessment of the functional impact of a stenosis in coronary heart disease. In this study, three models of artificial intelligence of varying degrees of complexity were compared to fractional flow reserve measurements. The three models are the multivariate polynomial regression, which is a statistical method used primarily for correlation; the feed-forward neural network; and the long short-term memory, which is a type of recurrent neural network that is suited to modelling sequences. The models were initially trained using a virtual patient database that was generated from a validated one-dimensional physics-based model. The feed-forward neural network performed the best for all test cases considered, which were a single vessel case from a virtual patient database, a multi-vessel network from a virtual patient database, and 25 clinically invasive fractional flow reserve measurements from real patients. The feed-forward neural network model achieved around 99% diagnostic accuracy in both tests involving virtual patients, and a respectable 72% diagnostic accuracy when compared to the invasive fractional flow reserve measurements. The multivariate polynomial regression model performed well in the single vessel case, but struggled on network cases as the variation of input features was much larger. The long short-term memory performed well for the single vessel cases, but tended to have a bias towards a positive fractional flow reserve prediction for the virtual multi-vessel case, and for the patient cases. Overall, the feed-forward neural network shows promise in successfully predicting fractional flow reserve in real patients, and could be a viable option if trained using a large enough data set of real patients.
published_date 2020-11-01T04:19:19Z
_version_ 1693001826369011712
score 10.780472