Blood fluid dynamics impacts vessel wall cells and tissue biomechanics, influencing thrombus formation and vessel wall remodelling. Accurate in vivo quantification can thus aid in understanding these mechanisms and patient stratification. Computational fluid dynamics (CFD) and 4D flow magnetic resonance imaging (MRI) are both used for this but have limitations: CFD involves assumptions and boundary condition simplifications, while 4D flow MRI suffers from low spatial resolution and noise. This study employs variational data assimilation to integrate CFD and 4D flow MRI, yielding a high-resolution, noise-free flow field closely aligned with 4D flow MRI velocity data. To enhance alignment, the optimal inlet velocity profile is determined iteratively via an incremental pressure correction scheme. Previously tested in simple synthetic geometries and later in a complex patient-specific abdominal aortic aneurysm, this approach demonstrates improved reliability in patient-specific haemodynamic evaluation.