amp;Ms and EMIDEC Challenges. STACOM 2020. Lecture Notes in Computer Science
4D flow magnetic resonance imaging (MRI) and in-silico simulations have seen widespread use in the characterization of blood flow patterns in the aorta and subsequent calibration of haemodynamic computational models. Computational Fluid Dynamics (CFD) simulations offer a complete overview on local haemodynamics but require patient-specific boundary conditions to provide realistic simulations. Despite the inherent low spatial resolution of 4D flow MRI near the boundaries, it can provide rich haemodynamic details to improve existing simulations. Unfortunately, very few works exist imaging the left atria (LA) with 4D flow MRI due to the acquisition and processing challenges associated to the low magnitude of velocities, the small size of the structure and the complexity of blood flow patterns, especially in pathologies such as atrial fibrillation (AF). The main goal of this study was to develop a computational pipeline to extract qualitative and quantitative indices of LA haemodynamics from 4D flow MRI to: assess differences between normal and AF left atria; and calibrate existing fluid models with improved boundary conditions. The preliminary results obtained in two cases demonstrate the potential of 4D flow MRI data to identify haemodynamic differences between healthy and AF left atria. Furthermore, it can help to bring flow computational simulations to a new level of realism, allowing more degrees of freedom to better capture the complexity of LA blood flow patterns.
Morales X, Mill J, Delso G, Loncaric F, Doltra A, Freixa X, Sitges M, Bijnens B, Camara O. 4D Flow Magnetic Resonance Imaging for Left Atrial Haemodynamic Characterization and Model Calibration. In: Puyol Anton E. et al.. Statistical Atlases and Computational Models of the Heart. M&Ms and EMIDEC Challenges. STACOM 2020. Lecture Notes in Computer Science. 1 ed. Springer; 2021. p. 156-165.