Evaluation of a benchtop MRI system for three-dimensional MR velocimetry in biofluidic flow

Abstract Accurate, non-invasive flow measurements are essential for validating computational models and investigating complex haemodynamic flows in biomedical applications. While magnetic resonance velocimetry (MRV) offers volumetric flow data, its utility is limited by high costs, infrastructure requirements, and clinical prioritisation. This study systematically investigates the feasibility of a compact 0.55 T benchtop MRI system with a 15 mm bore diameter for performing quantitative velocimetry in in vitro experimentation using three geometries of increasing complexity: a straight cylindrical pipe, a helical tube, and a patient-derived arteriovenous malformation (AVM) phantom. Flow measurements were evaluated against analytical solutions and computational fluid dynamics (CFD) reference simulations. The system successfully captured three-dimensional velocity fields across all geometries. Cylindrical pipe flow measurements achieved good quantitative agreement with theoretical predictions ( $$\delta =1.74\%$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>δ</mml:mi> <mml:mo>=</mml:mo> <mml:mn>1.74</mml:mn> <mml:mo>%</mml:mo> </mml:mrow> </mml:math> ), while helical flow imaging followed a similar trend ( $$\delta =8\%$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>δ</mml:mi> <mml:mo>=</mml:mo> <mml:mn>8</mml:mn> <mml:mo>%</mml:mo> </mml:mrow> </mml:math> ). The patient-derived flow phantom showed strong qualitative agreement with the CFD dataset with regard to flow structure and vector orientation. Bulk flow regions exhibited percentage velocity magnitude differences between 1 and $$3\%$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mn>3</mml:mn> <mml:mo>%</mml:mo> </mml:mrow> </mml:math> relative to the CFD reference dataset, with localised elevated errors of up to $$22\%$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mn>22</mml:mn> <mml:mo>%</mml:mo> </mml:mrow> </mml:math> in the low-velocity vortical flow, attributable to intravoxel velocity dispersion. The primary flow component yielded an RMSE of 16.12 mm/s and the overall velocity magnitude RMSE was 22.40 mm/s, representing $$8.1\%$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mn>8.1</mml:mn> <mml:mo>%</mml:mo> </mml:mrow> </mml:math> and $$11.2\%$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mn>11.2</mml:mn> <mml:mo>%</mml:mo> </mml:mrow> </mml:math> of the applied VENC, respectively. These results demonstrate that benchtop MRI systems can provide useful flow measurements for biofluidic studies. Although constrained by smaller bore sizes relative to clinical systems, benchtop scanners offer greater accessibility, lower cost, reduced infrastructure requirements, and non-invasive measurement capability, making them a promising alternative to clinical MRI and optical methods.