4DFlowNet: Super-Resolution 4D Flow MRI Using Deep Learning and Computational Fluid Dynamics
4D flow magnetic resonance imaging (MRI) is an emerging imaging technique where spatiotemporal 3D blood velocity can be captured with full volumetric coverage in a single non-invasive examination. This enables qualitative and quantitative analysis of hemodynamic flow parameters of the heart and grea...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2020-05-01
|
| Series: | Frontiers in Physics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fphy.2020.00138/full |
| _version_ | 1828766401284800512 |
|---|---|
| author | Edward Ferdian Avan Suinesiaputra Avan Suinesiaputra David J. Dubowitz Debbie Zhao Alan Wang Alan Wang Brett Cowan Brett Cowan Alistair A. Young Alistair A. Young |
| author_facet | Edward Ferdian Avan Suinesiaputra Avan Suinesiaputra David J. Dubowitz Debbie Zhao Alan Wang Alan Wang Brett Cowan Brett Cowan Alistair A. Young Alistair A. Young |
| author_sort | Edward Ferdian |
| collection | DOAJ |
| description | 4D flow magnetic resonance imaging (MRI) is an emerging imaging technique where spatiotemporal 3D blood velocity can be captured with full volumetric coverage in a single non-invasive examination. This enables qualitative and quantitative analysis of hemodynamic flow parameters of the heart and great vessels. An increase in the image resolution would provide more accuracy and allow better assessment of the blood flow, especially for patients with abnormal flows. However, this must be balanced with increasing imaging time. The recent success of deep learning in generating super resolution images shows promise for implementation in medical images. We utilized computational fluid dynamics simulations to generate fluid flow simulations and represent them as synthetic 4D flow MRI data. We built our training dataset to mimic actual 4D flow MRI data with its corresponding noise distribution. Our novel 4DFlowNet network was trained on this synthetic 4D flow data and was capable in producing noise-free super resolution 4D flow phase images with upsample factor of 2. We also tested the 4DFlowNet in actual 4D flow MR images of a phantom and normal volunteer data, and demonstrated comparable results with the actual flow rate measurements giving an absolute relative error of 0.6–5.8% and 1.1–3.8% in the phantom data and normal volunteer data, respectively. |
| first_indexed | 2024-12-11T07:05:10Z |
| format | Article |
| id | doaj.art-0485ced3b7a046efbff9bc36a07fcd01 |
| institution | Directory Open Access Journal |
| issn | 2296-424X |
| language | English |
| last_indexed | 2024-12-11T07:05:10Z |
| publishDate | 2020-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Physics |
| spelling | doaj.art-0485ced3b7a046efbff9bc36a07fcd012022-12-22T01:16:31ZengFrontiers Media S.A.Frontiers in Physics2296-424X2020-05-01810.3389/fphy.2020.001385335014DFlowNet: Super-Resolution 4D Flow MRI Using Deep Learning and Computational Fluid DynamicsEdward Ferdian0Avan Suinesiaputra1Avan Suinesiaputra2David J. Dubowitz3Debbie Zhao4Alan Wang5Alan Wang6Brett Cowan7Brett Cowan8Alistair A. Young9Alistair A. Young10Department of Anatomy and Medical Imaging, University of Auckland, Auckland, New ZealandDepartment of Anatomy and Medical Imaging, University of Auckland, Auckland, New ZealandSchool of Computing, University of Leeds, Leeds, United KingdomDepartment of Anatomy and Medical Imaging, University of Auckland, Auckland, New ZealandAuckland Bioengineering Institute, University of Auckland, Auckland, New ZealandDepartment of Anatomy and Medical Imaging, University of Auckland, Auckland, New ZealandAuckland Bioengineering Institute, University of Auckland, Auckland, New ZealandDepartment of Anatomy and Medical Imaging, University of Auckland, Auckland, New ZealandInstitute of Environmental Science and Research, Auckland, New ZealandDepartment of Anatomy and Medical Imaging, University of Auckland, Auckland, New ZealandDepartment of Biomedical Engineering, King's College London, London, United Kingdom4D flow magnetic resonance imaging (MRI) is an emerging imaging technique where spatiotemporal 3D blood velocity can be captured with full volumetric coverage in a single non-invasive examination. This enables qualitative and quantitative analysis of hemodynamic flow parameters of the heart and great vessels. An increase in the image resolution would provide more accuracy and allow better assessment of the blood flow, especially for patients with abnormal flows. However, this must be balanced with increasing imaging time. The recent success of deep learning in generating super resolution images shows promise for implementation in medical images. We utilized computational fluid dynamics simulations to generate fluid flow simulations and represent them as synthetic 4D flow MRI data. We built our training dataset to mimic actual 4D flow MRI data with its corresponding noise distribution. Our novel 4DFlowNet network was trained on this synthetic 4D flow data and was capable in producing noise-free super resolution 4D flow phase images with upsample factor of 2. We also tested the 4DFlowNet in actual 4D flow MR images of a phantom and normal volunteer data, and demonstrated comparable results with the actual flow rate measurements giving an absolute relative error of 0.6–5.8% and 1.1–3.8% in the phantom data and normal volunteer data, respectively.https://www.frontiersin.org/article/10.3389/fphy.2020.00138/full4D flow MRIsuper resolution networkSRResNetdeep learningcomputational fluid dynamicsCFD |
| spellingShingle | Edward Ferdian Avan Suinesiaputra Avan Suinesiaputra David J. Dubowitz Debbie Zhao Alan Wang Alan Wang Brett Cowan Brett Cowan Alistair A. Young Alistair A. Young 4DFlowNet: Super-Resolution 4D Flow MRI Using Deep Learning and Computational Fluid Dynamics Frontiers in Physics 4D flow MRI super resolution network SRResNet deep learning computational fluid dynamics CFD |
| title | 4DFlowNet: Super-Resolution 4D Flow MRI Using Deep Learning and Computational Fluid Dynamics |
| title_full | 4DFlowNet: Super-Resolution 4D Flow MRI Using Deep Learning and Computational Fluid Dynamics |
| title_fullStr | 4DFlowNet: Super-Resolution 4D Flow MRI Using Deep Learning and Computational Fluid Dynamics |
| title_full_unstemmed | 4DFlowNet: Super-Resolution 4D Flow MRI Using Deep Learning and Computational Fluid Dynamics |
| title_short | 4DFlowNet: Super-Resolution 4D Flow MRI Using Deep Learning and Computational Fluid Dynamics |
| title_sort | 4dflownet super resolution 4d flow mri using deep learning and computational fluid dynamics |
| topic | 4D flow MRI super resolution network SRResNet deep learning computational fluid dynamics CFD |
| url | https://www.frontiersin.org/article/10.3389/fphy.2020.00138/full |
| work_keys_str_mv | AT edwardferdian 4dflownetsuperresolution4dflowmriusingdeeplearningandcomputationalfluiddynamics AT avansuinesiaputra 4dflownetsuperresolution4dflowmriusingdeeplearningandcomputationalfluiddynamics AT avansuinesiaputra 4dflownetsuperresolution4dflowmriusingdeeplearningandcomputationalfluiddynamics AT davidjdubowitz 4dflownetsuperresolution4dflowmriusingdeeplearningandcomputationalfluiddynamics AT debbiezhao 4dflownetsuperresolution4dflowmriusingdeeplearningandcomputationalfluiddynamics AT alanwang 4dflownetsuperresolution4dflowmriusingdeeplearningandcomputationalfluiddynamics AT alanwang 4dflownetsuperresolution4dflowmriusingdeeplearningandcomputationalfluiddynamics AT brettcowan 4dflownetsuperresolution4dflowmriusingdeeplearningandcomputationalfluiddynamics AT brettcowan 4dflownetsuperresolution4dflowmriusingdeeplearningandcomputationalfluiddynamics AT alistairayoung 4dflownetsuperresolution4dflowmriusingdeeplearningandcomputationalfluiddynamics AT alistairayoung 4dflownetsuperresolution4dflowmriusingdeeplearningandcomputationalfluiddynamics |