Abstract
Blood flow metrics obtained with four-dimensional (4D) flow phase contrast (PC) magnetic resonance imaging (MRI) can be of great value in clinical and experimental cerebrovascular analysis. However, limitations in both quantitative and qualitative analyses can result from errors inherent to PC MRI. One method that excels in creating low-error, physics-based, velocity fields is computational fluid dynamics (CFD). Augmentation of cerebral 4D flow MRI data with CFD-informed neural networks may provide a method to produce highly accurate physiological flow fields. In this preliminary study, the potential utility of such a method was demonstrated by using high resolution patient-specific CFD data to train a convolutional neural network, and then using the trained network to enhance MRI-derived velocity fields in cerebral blood vessel data sets. Through testing on simulated images, phantom data, and cerebrovascular 4D flow data from 20 patients, the trained network successfully de-noised flow images, decreased velocity error, and enhanced near-vessel-wall velocity quantification and visualization. Such image enhancement can improve experimental and clinical qualitative and quantitative cerebrovascular PC MRI analysis.
Highlights
A number of studies have been aimed at taking advantage of the synergistic strengths of phase contrast (PC) magnetic resonance imaging (MRI) and computational simulation to address current flow characterization limitations in a variety of vascular territories
Velocity error was reduced from 20% in the corrupted velocity field to 8% in the post-training velocity field when the original computational fluid dynamics (CFD) velocity field was taken as the ground truth
The potential utility of such a method was demonstrated by using high resolution patient-specific CFD data to train a neural network, and using the trained network to enhance MRI-derived cerebrovascular velocity fields
Summary
Data from CFD and PC MRI are often compared to move towards validation of quantitative and qualitative flow characterization[23] Building on these advancements, a more robust flow physics-based method that allows for retrospective correction of PC MRI could prove highly beneficial for experimental and clinical cardiovascular flow analysis with MRI. Application of machine learning methods to improve the velocity fields of phase contrast MRI has been limited outside of these recent studies, despite its potential for cerebrovascular flow quantification improvement. The purpose of this work was to develop a robust machine learning paradigm which fuses information from 4D flow MRI and CFD using supervised learning, to provide high resolution, physics-based, patient-specific flow fields in cerebrovascular anatomy
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
