Loading…

Augmentation of experimentally obtained flow fields by means of Physics Informed Neural Networks (PINN) demonstrated on aneurysm flow

Biofluid mechanics play an important role in the study of the mechanism of cardiovascular diseases and in the development of new implants. For the assessment of hydrodynamic parameters, experimental methods as well as in-silico approaches can be used, such as particle image velocimetry (PIV) and Dee...

Full description

Saved in:
Bibliographic Details
Published in:Current directions in biomedical engineering 2023-09, Vol.9 (1), p.519-523
Main Authors: Oldenburg, Jan, Borowski, Finja, Wollenberg, Wiebke, Öner, Alper, Schmitz, Klaus-Peter, Stiehm, Michael
Format: Article
Language:English
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 523
container_issue 1
container_start_page 519
container_title Current directions in biomedical engineering
container_volume 9
creator Oldenburg, Jan
Borowski, Finja
Wollenberg, Wiebke
Öner, Alper
Schmitz, Klaus-Peter
Stiehm, Michael
description Biofluid mechanics play an important role in the study of the mechanism of cardiovascular diseases and in the development of new implants. For the assessment of hydrodynamic parameters, experimental methods as well as in-silico approaches can be used, such as particle image velocimetry (PIV) and Deep Learning, respectively. Challenges for PIV are the optical access to the region of interest, and time consumption for measuring and post-processing analysis in particular for three dimensional flow. To overcome these limitations state-of-the-art deep learning algorithms could be utilized to augment spatially coarse resolved flow fields. In this study, we demonstrate the use of Physics Informed Neural Networks (PINN) to augment PIV measurement data. To demonstrate a combined workflow, we investigate the flow of a Newtonian fluid through a simplified aneurysm under laminar conditions. Generation of synthetic PIV particle images of a single measurement plane and the corresponding PIV vector calculations were performed as the basis for the PINN algorithm. Based on the Navier-Stokes equations the PINN reconstructs the entire 3D flow field and pressure distribution inside the aneurysm. We observed qualitative agreements between ground through data and PINN predictions. Nevertheless, there are substantial differences in the quantitative, locally resolved comparison of the flow metrics, despite the generally tendency for the PINN algorithm to correctly augment the flow field.
doi_str_mv 10.1515/cdbme-2023-1130
format article
fullrecord <record><control><sourceid>walterdegruyter_doaj_</sourceid><recordid>TN_cdi_doaj_primary_oai_doaj_org_article_9060658c87e040939808ac7478888666</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><doaj_id>oai_doaj_org_article_9060658c87e040939808ac7478888666</doaj_id><sourcerecordid>10_1515_cdbme_2023_113091519</sourcerecordid><originalsourceid>FETCH-LOGICAL-c2709-d391d97ba86a36ce644ac6412c6e54da4605ce475b8133256501aff65d7e50683</originalsourceid><addsrcrecordid>eNp1UcuO1DAQjBBIrJY9c_URDmHbsd1JJC6rFY-RVsMe4Gx17M6QIYlXdkZDPoD_xjODEBf60q1SVbVKVRSvJbyTRppb57uJywoqVUqp4FlxVSnUpTGgn_9zvyxuUtoDgMQKsVFXxa-7w27ieaFlCLMIveCfTxyHMzSOqwjdQsPMXvRjOIp-4NEn0a1iYprTif_4fU2DS2Iz9yFOmbjlQ6Qxr-UY4o8k3jxuttu3wvMU5rREWjInv6I589Y0nY1fFS96GhPf_NnXxbePH77efy4fvnza3N89lK6qoS29aqVv644aJIWOUWtyqGXlkI32pBGMY12brpFKVQYNSOp7NL5mAznvdbG5-PpAe_uUc1JcbaDBnoEQd5biMriRbQsIaBrX1AwaWtU20JCrdd3kQcTsdXvxcjGkFLn_6yfBnkqx51LsqRR7KiUr3l8URxoXjp538bDmw-7DIc459_-UbQZb9RvR8pQt</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Augmentation of experimentally obtained flow fields by means of Physics Informed Neural Networks (PINN) demonstrated on aneurysm flow</title><source>Sciendo (De Gruyter) Open Access Journals</source><creator>Oldenburg, Jan ; Borowski, Finja ; Wollenberg, Wiebke ; Öner, Alper ; Schmitz, Klaus-Peter ; Stiehm, Michael</creator><creatorcontrib>Oldenburg, Jan ; Borowski, Finja ; Wollenberg, Wiebke ; Öner, Alper ; Schmitz, Klaus-Peter ; Stiehm, Michael</creatorcontrib><description>Biofluid mechanics play an important role in the study of the mechanism of cardiovascular diseases and in the development of new implants. For the assessment of hydrodynamic parameters, experimental methods as well as in-silico approaches can be used, such as particle image velocimetry (PIV) and Deep Learning, respectively. Challenges for PIV are the optical access to the region of interest, and time consumption for measuring and post-processing analysis in particular for three dimensional flow. To overcome these limitations state-of-the-art deep learning algorithms could be utilized to augment spatially coarse resolved flow fields. In this study, we demonstrate the use of Physics Informed Neural Networks (PINN) to augment PIV measurement data. To demonstrate a combined workflow, we investigate the flow of a Newtonian fluid through a simplified aneurysm under laminar conditions. Generation of synthetic PIV particle images of a single measurement plane and the corresponding PIV vector calculations were performed as the basis for the PINN algorithm. Based on the Navier-Stokes equations the PINN reconstructs the entire 3D flow field and pressure distribution inside the aneurysm. We observed qualitative agreements between ground through data and PINN predictions. Nevertheless, there are substantial differences in the quantitative, locally resolved comparison of the flow metrics, despite the generally tendency for the PINN algorithm to correctly augment the flow field.</description><identifier>ISSN: 2364-5504</identifier><identifier>EISSN: 2364-5504</identifier><identifier>DOI: 10.1515/cdbme-2023-1130</identifier><language>eng</language><publisher>De Gruyter</publisher><subject>Aneurysm ; Deep Learning ; Flow Field Augmentation ; Particle Image Velocimetry ; Physics Informed Neural Networks</subject><ispartof>Current directions in biomedical engineering, 2023-09, Vol.9 (1), p.519-523</ispartof><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.degruyter.com/document/doi/10.1515/cdbme-2023-1130/pdf$$EPDF$$P50$$Gwalterdegruyter$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://www.degruyter.com/document/doi/10.1515/cdbme-2023-1130/html$$EHTML$$P50$$Gwalterdegruyter$$Hfree_for_read</linktohtml><link.rule.ids>314,777,781,27905,27906,66907,68691</link.rule.ids></links><search><creatorcontrib>Oldenburg, Jan</creatorcontrib><creatorcontrib>Borowski, Finja</creatorcontrib><creatorcontrib>Wollenberg, Wiebke</creatorcontrib><creatorcontrib>Öner, Alper</creatorcontrib><creatorcontrib>Schmitz, Klaus-Peter</creatorcontrib><creatorcontrib>Stiehm, Michael</creatorcontrib><title>Augmentation of experimentally obtained flow fields by means of Physics Informed Neural Networks (PINN) demonstrated on aneurysm flow</title><title>Current directions in biomedical engineering</title><description>Biofluid mechanics play an important role in the study of the mechanism of cardiovascular diseases and in the development of new implants. For the assessment of hydrodynamic parameters, experimental methods as well as in-silico approaches can be used, such as particle image velocimetry (PIV) and Deep Learning, respectively. Challenges for PIV are the optical access to the region of interest, and time consumption for measuring and post-processing analysis in particular for three dimensional flow. To overcome these limitations state-of-the-art deep learning algorithms could be utilized to augment spatially coarse resolved flow fields. In this study, we demonstrate the use of Physics Informed Neural Networks (PINN) to augment PIV measurement data. To demonstrate a combined workflow, we investigate the flow of a Newtonian fluid through a simplified aneurysm under laminar conditions. Generation of synthetic PIV particle images of a single measurement plane and the corresponding PIV vector calculations were performed as the basis for the PINN algorithm. Based on the Navier-Stokes equations the PINN reconstructs the entire 3D flow field and pressure distribution inside the aneurysm. We observed qualitative agreements between ground through data and PINN predictions. Nevertheless, there are substantial differences in the quantitative, locally resolved comparison of the flow metrics, despite the generally tendency for the PINN algorithm to correctly augment the flow field.</description><subject>Aneurysm</subject><subject>Deep Learning</subject><subject>Flow Field Augmentation</subject><subject>Particle Image Velocimetry</subject><subject>Physics Informed Neural Networks</subject><issn>2364-5504</issn><issn>2364-5504</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>DOA</sourceid><recordid>eNp1UcuO1DAQjBBIrJY9c_URDmHbsd1JJC6rFY-RVsMe4Gx17M6QIYlXdkZDPoD_xjODEBf60q1SVbVKVRSvJbyTRppb57uJywoqVUqp4FlxVSnUpTGgn_9zvyxuUtoDgMQKsVFXxa-7w27ieaFlCLMIveCfTxyHMzSOqwjdQsPMXvRjOIp-4NEn0a1iYprTif_4fU2DS2Iz9yFOmbjlQ6Qxr-UY4o8k3jxuttu3wvMU5rREWjInv6I589Y0nY1fFS96GhPf_NnXxbePH77efy4fvnza3N89lK6qoS29aqVv644aJIWOUWtyqGXlkI32pBGMY12brpFKVQYNSOp7NL5mAznvdbG5-PpAe_uUc1JcbaDBnoEQd5biMriRbQsIaBrX1AwaWtU20JCrdd3kQcTsdXvxcjGkFLn_6yfBnkqx51LsqRR7KiUr3l8URxoXjp538bDmw-7DIc459_-UbQZb9RvR8pQt</recordid><startdate>20230901</startdate><enddate>20230901</enddate><creator>Oldenburg, Jan</creator><creator>Borowski, Finja</creator><creator>Wollenberg, Wiebke</creator><creator>Öner, Alper</creator><creator>Schmitz, Klaus-Peter</creator><creator>Stiehm, Michael</creator><general>De Gruyter</general><scope>AAYXX</scope><scope>CITATION</scope><scope>DOA</scope></search><sort><creationdate>20230901</creationdate><title>Augmentation of experimentally obtained flow fields by means of Physics Informed Neural Networks (PINN) demonstrated on aneurysm flow</title><author>Oldenburg, Jan ; Borowski, Finja ; Wollenberg, Wiebke ; Öner, Alper ; Schmitz, Klaus-Peter ; Stiehm, Michael</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c2709-d391d97ba86a36ce644ac6412c6e54da4605ce475b8133256501aff65d7e50683</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Aneurysm</topic><topic>Deep Learning</topic><topic>Flow Field Augmentation</topic><topic>Particle Image Velocimetry</topic><topic>Physics Informed Neural Networks</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Oldenburg, Jan</creatorcontrib><creatorcontrib>Borowski, Finja</creatorcontrib><creatorcontrib>Wollenberg, Wiebke</creatorcontrib><creatorcontrib>Öner, Alper</creatorcontrib><creatorcontrib>Schmitz, Klaus-Peter</creatorcontrib><creatorcontrib>Stiehm, Michael</creatorcontrib><collection>CrossRef</collection><collection>Open Access: DOAJ - Directory of Open Access Journals</collection><jtitle>Current directions in biomedical engineering</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Oldenburg, Jan</au><au>Borowski, Finja</au><au>Wollenberg, Wiebke</au><au>Öner, Alper</au><au>Schmitz, Klaus-Peter</au><au>Stiehm, Michael</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Augmentation of experimentally obtained flow fields by means of Physics Informed Neural Networks (PINN) demonstrated on aneurysm flow</atitle><jtitle>Current directions in biomedical engineering</jtitle><date>2023-09-01</date><risdate>2023</risdate><volume>9</volume><issue>1</issue><spage>519</spage><epage>523</epage><pages>519-523</pages><issn>2364-5504</issn><eissn>2364-5504</eissn><abstract>Biofluid mechanics play an important role in the study of the mechanism of cardiovascular diseases and in the development of new implants. For the assessment of hydrodynamic parameters, experimental methods as well as in-silico approaches can be used, such as particle image velocimetry (PIV) and Deep Learning, respectively. Challenges for PIV are the optical access to the region of interest, and time consumption for measuring and post-processing analysis in particular for three dimensional flow. To overcome these limitations state-of-the-art deep learning algorithms could be utilized to augment spatially coarse resolved flow fields. In this study, we demonstrate the use of Physics Informed Neural Networks (PINN) to augment PIV measurement data. To demonstrate a combined workflow, we investigate the flow of a Newtonian fluid through a simplified aneurysm under laminar conditions. Generation of synthetic PIV particle images of a single measurement plane and the corresponding PIV vector calculations were performed as the basis for the PINN algorithm. Based on the Navier-Stokes equations the PINN reconstructs the entire 3D flow field and pressure distribution inside the aneurysm. We observed qualitative agreements between ground through data and PINN predictions. Nevertheless, there are substantial differences in the quantitative, locally resolved comparison of the flow metrics, despite the generally tendency for the PINN algorithm to correctly augment the flow field.</abstract><pub>De Gruyter</pub><doi>10.1515/cdbme-2023-1130</doi><tpages>5</tpages><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 2364-5504
ispartof Current directions in biomedical engineering, 2023-09, Vol.9 (1), p.519-523
issn 2364-5504
2364-5504
language eng
recordid cdi_doaj_primary_oai_doaj_org_article_9060658c87e040939808ac7478888666
source Sciendo (De Gruyter) Open Access Journals
subjects Aneurysm
Deep Learning
Flow Field Augmentation
Particle Image Velocimetry
Physics Informed Neural Networks
title Augmentation of experimentally obtained flow fields by means of Physics Informed Neural Networks (PINN) demonstrated on aneurysm flow
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-19T12%3A44%3A45IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-walterdegruyter_doaj_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Augmentation%20of%20experimentally%20obtained%20flow%20fields%20by%20means%20of%20Physics%20Informed%20Neural%20Networks%20(PINN)%20demonstrated%20on%20aneurysm%20flow&rft.jtitle=Current%20directions%20in%20biomedical%20engineering&rft.au=Oldenburg,%20Jan&rft.date=2023-09-01&rft.volume=9&rft.issue=1&rft.spage=519&rft.epage=523&rft.pages=519-523&rft.issn=2364-5504&rft.eissn=2364-5504&rft_id=info:doi/10.1515/cdbme-2023-1130&rft_dat=%3Cwalterdegruyter_doaj_%3E10_1515_cdbme_2023_113091519%3C/walterdegruyter_doaj_%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c2709-d391d97ba86a36ce644ac6412c6e54da4605ce475b8133256501aff65d7e50683%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true