(E-178) Computational modeling of left ventricular flow using CMR-derived four-dimensional wall motion

Introduction:: Introduction: Simulations of blood flow in the left ventricle (LV) have been shown to provide valuable insight into the pathophysiology of cardiovascular diseases (CVDs). Recently, the analysis of flow interactions with the endocardial wall using medical imaging data has been used to detect regional abnormal function in the LV. The excellent soft-tissue contrast and superior temporal resolution offered by cardiac magnetic resonance (CMR) images have shown promise in evaluating LV deformation [1, 2]. Most CMR-based studies continue to utilize the imaging data for only patient-specific reconstruction and quantification of cardiac strains. In addition to quantifying cardiac motion, the CMR images offer the possibility of developing an imaging-based model of ventricular fluid structure interactions (FSI) capturing the interactions between LV kinematics and blood flow. In this work, we implement a non-rigid image registration framework to evaluate the structural deformation of the myocardial wall. Our method extends upon previous CMR-based methods, with the non-rigid estimation of 4D myocardial deformation eliminating the need for involved myocardial material modeling to capture LV wall FSI. In addition, phase contrast (PC-)CMR images were used , in conjunction with standard cine CMR, to measure spatiotemporal blood flow velocity at the aortic, and mitral valves[3].


Materials and Methods:: Materials and Methods: Image registration: The short-axis (SA) CMR images were used to identify end-diastolic (ED) and end-systolic (ES) time points, and data points (pixels) were longitudinally up-sampled through cubic interpolation to reduce through-plane anisotropy. An in-house non-rigid image registration framework was implemented over the resulting image stack to evaluate deformation[4]. Also, “through-plane” displacements were calculated via longitudinal pixel tracking across an image stack of multiple SA slices to estimate the wall deformation. Following image registration, the LV was segmented over a single cardiac cycle at ED time point. Flow simulation: For the fluid–structure interaction modeling, the boundary motion was prescribed on the LV wall using a user-defined function in Ansys Fluent. A dynamic remeshing was implemented and using high-fidelity three-dimensional wall displacement over the entire cardiac cycle. The unsteady forms of the continuity and momentum equations were solved, and the relative velocity was used to recapture the moving mesh in the ALE formulation of Navier–Stokes equations. The PC-CMR images were used to determine the rate of opening of the mitral valve, and also to validate the inlet and outlet flow velocity. A summary of the described pipeline is presented in Fig. 1.


Results, Conclusions, and Discussions:: Results and Discussions: Analyses were performed for two cardiac cycles, and the preliminary results show a consistency with existing literature. Organ-level flow patterns were recapitulated, with the fluid flow driven primarily through the CMR-derived wall motion. At the onset of ED, with aortic and mitral valves closed, the stagnation of blood flow in the chamber was captured. The generation of a vortex ring is a crucial aspect of LV inflow, and our investigations will be directed towards the recapitulation of this phenomenon. Also, the boundary conditions reinforced through PC-CMR images will allow the extension of our pipeline to obtain the maximum flow velocity of the impinging jet of diastolic flow through the mitral valve.

Conclusions: An FSI model was presented to capture the hemodynamic characteristics of blood flow in a deforming LV. The methodology was established using standard and noninvasive CMR and PC-CMR images, which enhances the translational potential of the method for future studies. Furthermore, the proposed method holds the promise to replicate the complex flow patterns in the LV without the need for conventional fluid-solid formulations. This integration of standard imaging practice with in-silico simulations is expected to provide valuable tool for patient-specific design of cardiac interventions in CVDs.


Acknowledgements (Optional): :

References (Optional): : References:

1. Pedrizzetti, G., et al., Principles of cardiovascular magnetic resonance feature tracking and echocardiographic speckle tracking for informed clinical use. Journal of cardiovascular magnetic resonance, 2016. 18(1): p. 1-12.

2. Thomas, D., et al., Quantitative assessment of regional myocardial function in a rat model of myocardial infarction using tagged MRI. Magnetic Resonance Materials in Physics, Biology and Medicine, 2004. 17: p. 179-187.

3. Markl, M., P.J. Kilner, and T. Ebbers, Comprehensive 4D velocity mapping of the heart and great vessels by cardiovascular magnetic resonance. Journal of Cardiovascular Magnetic Resonance, 2011. 13: p. 1-22.

4. Keshavarzian, M., et al. An image registration framework to estimate 3d myocardial strains from cine cardiac mri in mice. in Functional Imaging and Modeling of the Heart: 11th International Conference, FIMH 2021, Stanford, CA, USA, June 21-25, 2021, Proceedings. 2021. Springer.