Validation of 4D Flow MRI for Jet Flow Characterization Under Steady Conditions

Accurate assessment of jet flow dynamics is essential for diagnosing valvular heart diseases. Although 4D Flow MRI enables three-dimensional visualization of cardiovascular flow, its precision is often limited by resolution-dependent artifacts. This study aims to develop a reliable method for observing jet blood flow by comparing 4D Flow MRI with Particle Image Velocimetry (PIV). In-vitro experiments were conducted using three jet models under steady flow conditions. Velocity fields were obtained using both PIV and 4D Flow MRI at voxel sizes of 1 mm, 2 mm, and 3 mm. An FFT-based low-pass filter was applied to MRI data to improve the resolution and suppress noise. Jet features such as centerline, length, and thickness were extracted from both modalities and evaluated using correlation metrics, Bland-Altman analysis, and spatial error mapping. The results showed increasing deviation from PIV measurements as voxel size increased, especially in asymmetric jet models. While filtering and reconstruction techniques improved MRI performance at finer resolutions, significant information loss remained at coarser resolutions. These findings emphasize the importance of high-resolution imaging and resolution-aware post-processing to improve the reliability of MRI-based characterization of jet flows. The comparative analysis with PIV provides insight into the limitations and potential corrections for MRI in cardiovascular flow assessment, which is crucial for accurate diagnosis and improved treatment planning in clinical cardiology.