Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields.

Computational fluid dynamics (CFD) can be used to simulate vascular haemodynamics and analyse potential treatment options. CFD has shown to be beneficial in improving patient outcomes. However, the implementation of CFD for routine clinical use is yet to be realised. Barriers for CFD include high co...

Full description

Saved in:
Bibliographic Details
Main Authors: Endrit Pajaziti, Javier Montalt-Tordera, Claudio Capelli, Raphaël Sivera, Emilie Sauvage, Michael Quail, Silvia Schievano, Vivek Muthurangu
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2023-04-01
Series:PLoS Computational Biology
Online Access:https://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1011055&type=printable
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849470026261725184
author Endrit Pajaziti
Javier Montalt-Tordera
Claudio Capelli
Raphaël Sivera
Emilie Sauvage
Michael Quail
Silvia Schievano
Vivek Muthurangu
author_facet Endrit Pajaziti
Javier Montalt-Tordera
Claudio Capelli
Raphaël Sivera
Emilie Sauvage
Michael Quail
Silvia Schievano
Vivek Muthurangu
author_sort Endrit Pajaziti
collection DOAJ
description Computational fluid dynamics (CFD) can be used to simulate vascular haemodynamics and analyse potential treatment options. CFD has shown to be beneficial in improving patient outcomes. However, the implementation of CFD for routine clinical use is yet to be realised. Barriers for CFD include high computational resources, specialist experience needed for designing simulation set-ups, and long processing times. The aim of this study was to explore the use of machine learning (ML) to replicate conventional aortic CFD with automatic and fast regression models. Data used to train/test the model consisted of 3,000 CFD simulations performed on synthetically generated 3D aortic shapes. These subjects were generated from a statistical shape model (SSM) built on real patient-specific aortas (N = 67). Inference performed on 200 test shapes resulted in average errors of 6.01% ±3.12 SD and 3.99% ±0.93 SD for pressure and velocity, respectively. Our ML-based models performed CFD in ∼0.075 seconds (4,000x faster than the solver). This proof-of-concept study shows that results from conventional vascular CFD can be reproduced using ML at a much faster rate, in an automatic process, and with reasonable accuracy.
format Article
id doaj-art-88220f5e6c8f43c681a9f8ed3313e324
institution Kabale University
issn 1553-734X
1553-7358
language English
publishDate 2023-04-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS Computational Biology
spelling doaj-art-88220f5e6c8f43c681a9f8ed3313e3242025-08-20T03:25:16ZengPublic Library of Science (PLoS)PLoS Computational Biology1553-734X1553-73582023-04-01194e101105510.1371/journal.pcbi.1011055Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields.Endrit PajazitiJavier Montalt-TorderaClaudio CapelliRaphaël SiveraEmilie SauvageMichael QuailSilvia SchievanoVivek MuthuranguComputational fluid dynamics (CFD) can be used to simulate vascular haemodynamics and analyse potential treatment options. CFD has shown to be beneficial in improving patient outcomes. However, the implementation of CFD for routine clinical use is yet to be realised. Barriers for CFD include high computational resources, specialist experience needed for designing simulation set-ups, and long processing times. The aim of this study was to explore the use of machine learning (ML) to replicate conventional aortic CFD with automatic and fast regression models. Data used to train/test the model consisted of 3,000 CFD simulations performed on synthetically generated 3D aortic shapes. These subjects were generated from a statistical shape model (SSM) built on real patient-specific aortas (N = 67). Inference performed on 200 test shapes resulted in average errors of 6.01% ±3.12 SD and 3.99% ±0.93 SD for pressure and velocity, respectively. Our ML-based models performed CFD in ∼0.075 seconds (4,000x faster than the solver). This proof-of-concept study shows that results from conventional vascular CFD can be reproduced using ML at a much faster rate, in an automatic process, and with reasonable accuracy.https://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1011055&type=printable
spellingShingle Endrit Pajaziti
Javier Montalt-Tordera
Claudio Capelli
Raphaël Sivera
Emilie Sauvage
Michael Quail
Silvia Schievano
Vivek Muthurangu
Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields.
PLoS Computational Biology
title Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields.
title_full Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields.
title_fullStr Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields.
title_full_unstemmed Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields.
title_short Shape-driven deep neural networks for fast acquisition of aortic 3D pressure and velocity flow fields.
title_sort shape driven deep neural networks for fast acquisition of aortic 3d pressure and velocity flow fields
url https://journals.plos.org/ploscompbiol/article/file?id=10.1371/journal.pcbi.1011055&type=printable
work_keys_str_mv AT endritpajaziti shapedrivendeepneuralnetworksforfastacquisitionofaortic3dpressureandvelocityflowfields
AT javiermontalttordera shapedrivendeepneuralnetworksforfastacquisitionofaortic3dpressureandvelocityflowfields
AT claudiocapelli shapedrivendeepneuralnetworksforfastacquisitionofaortic3dpressureandvelocityflowfields
AT raphaelsivera shapedrivendeepneuralnetworksforfastacquisitionofaortic3dpressureandvelocityflowfields
AT emiliesauvage shapedrivendeepneuralnetworksforfastacquisitionofaortic3dpressureandvelocityflowfields
AT michaelquail shapedrivendeepneuralnetworksforfastacquisitionofaortic3dpressureandvelocityflowfields
AT silviaschievano shapedrivendeepneuralnetworksforfastacquisitionofaortic3dpressureandvelocityflowfields
AT vivekmuthurangu shapedrivendeepneuralnetworksforfastacquisitionofaortic3dpressureandvelocityflowfields