Super-resolution 4D flow MRI to quantify aortic regurgitation using computational fluid dynamics and deep learning

Changes in cardiovascular hemodynamics are closely related to the development of aortic regurgitation (AR), a type of valvular heart disease. Metrics derived from blood flows are used to indicate AR onset and evaluate its severity. These metrics can be non-invasively obtained using four-dimensional...

Full description

Saved in:
Bibliographic Details
Published in:The international journal of cardiovascular imaging 2023-06, Vol.39 (6), p.1189-1202
Main Authors: Long, Derek, McMurdo, Cameron, Ferdian, Edward, Mauger, Charlène A., Marlevi, David, Nash, Martyn P., Young, Alistair A.
Format: Article
Language:English
Subjects:
Aorta
Aortic Valve Insufficiency - diagnostic imaging
Blood flow
Blood Flow Velocity - physiology
Cardiac Imaging
Cardiology
Cardiovascular diseases
Computational fluid dynamics
Computer applications
Deep Learning
Fluid dynamics
Heart diseases
Hemodynamics
Humans
Hydrodynamics
Image resolution
Imaging
Imaging, Three-Dimensional - methods
Magnetic resonance imaging
Magnetic Resonance Imaging - methods
Medical imaging
Medicin och hälsovetenskap
Medicine
Medicine & Public Health
Neural networks
Original Paper
Predictive Value of Tests
Radiology
Regurgitation
Spatial discrimination
Spatial resolution
Velocity errors
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Changes in cardiovascular hemodynamics are closely related to the development of aortic regurgitation (AR), a type of valvular heart disease. Metrics derived from blood flows are used to indicate AR onset and evaluate its severity. These metrics can be non-invasively obtained using four-dimensional (4D) flow magnetic resonance imaging (MRI), where accuracy is primarily dependent on spatial resolution. However, insufficient resolution often results from limitations in 4D flow MRI and complex aortic regurgitation hemodynamics. To address this, computational fluid dynamics simulations were transformed into synthetic 4D flow MRI data and used to train a variety of neural networks. These networks generated super-resolution, full-field phase images with an upsample factor of 4. Results showed decreased velocity error, high structural similarity scores, and improved learning capabilities from previous work. Further validation was performed on two sets of in vivo 4D flow MRI data and demonstrated success in de-noising flow images. This approach presents an opportunity to comprehensively analyse AR hemodynamics in a non-invasive manner.
ISSN:1875-8312
1569-5794
1875-8312
1573-0743
DOI:10.1007/s10554-023-02815-z