Refining wall shear stress measurements in four-dimensional flow magnetic resonance imaging through intra-voxel wall boundary consideration

Accurately estimating wall shear stress (WSS) is crucial for understanding the mechanisms and diagnosing cardiovascular diseases. Although four-dimensional flow magnetic resonance imaging (4D flow MRI) is a valuable tool for quantifying WSS, it is often affected by partial volume effects (PVE), leading to potential inaccuracies in WSS estimation. In this study, we propose a novel approach to enhance WSS estimation by addressing PVE in 4D flow MRI measurements. Our method involves precisely locating the vessel wall boundary within a voxel adjacent to the wall, leveraging a linear correlation between the magnitude of the magnetic resonance signal and the fraction of fluid occupancy in the voxel. The proposed method is evaluated by measuring laminar circular tubing flow with a 8 mm diameter and in vivo aortic data. Through our experiments, we demonstrate a noticeable improvement in the accuracy of WSS estimation when PVE is considered. The root mean square error of the measured WSS to the theoretical value decreases by approximately 40% with a higher Reynolds resolution (i.e., lower value). Additionally, the PVE consideration method shows about a 7% lower WSS median relative to without PVE consideration for aortic data. These findings emphasize the importance of accounting for PVE and striving for higher Reynolds resolutions in 4D flow MRI to achieve precise WSS estimation, ultimately benefiting clinical research and advancing the application of WSS in cardiovascular disease diagnosis.