Abstract
BackgroundThe development of clinically applicable fluid-structure interaction (FSI) models of the left heart is inherently challenging when using in vivo cardiovascular magnetic resonance (CMR) data for validation, due to the lack of a well-controlled system where detailed measurements of the ventricular wall motion and flow field are available a priori. The purpose of this study was to (a) develop a clinically relevant, CMR-compatible left heart physical model; and (b) compare the left ventricular (LV) volume reconstructions and hemodynamic data obtained using CMR to laboratory-based experimental modalities.MethodsThe LV was constructed from optically clear flexible silicone rubber. The geometry was based off a healthy patient’s LV geometry during peak systole. The LV phantom was attached to a left heart simulator consisting of an aorta, atrium, and systemic resistance and compliance elements. Experiments were conducted for heart rate of 70 bpm. Wall motion measurements were obtained using high speed stereo-photogrammetry (SP) and cine-CMR, while flow field measurements were obtained using digital particle image velocimetry (DPIV) and phase-contrast magnetic resonance (PC-CMR).ResultsThe model reproduced physiologically accurate hemodynamics (aortic pressure = 120/80 mmHg; cardiac output = 3.5 L/min). DPIV and PC-CMR results of the center plane flow within the ventricle matched, both qualitatively and quantitatively, with flow from the atrium into the LV having a velocity of about 1.15 m/s for both modalities. The normalized LV volume through the cardiac cycle computed from CMR data matched closely to that from SP. The mean difference between CMR and SP was 5.5 ± 3.7 %.ConclusionsThe model presented here can thus be used for the purposes of: (a) acquiring CMR data for validation of FSI simulations, (b) determining accuracy of cine-CMR reconstruction methods, and (c) conducting investigations of the effects of altering anatomical variables on LV function under normal and disease conditions.
Highlights
The development of clinically applicable fluid-structure interaction (FSI) models of the left heart is inherently challenging when using in vivo cardiovascular magnetic resonance (CMR) data for validation, due to the lack of a well-controlled system where detailed measurements of the ventricular wall motion and flow field are available a priori
LV physical model Design and construction The left ventricular geometry was generated in SolidworksTM (Dassault Systèmes Solidworks Corporation, Waltham, MA, USA) by constructing a series of concentric ellipses that were fit to LV endocardial borders traced on 5 cine steady-state free procession (SSFP), short axis slices at peak systolic phase of the cardiac cycle acquired in a healthy subject’s (Fig. 1)
The results of this study demonstrated the applicability of an CMR-compatible LV physical model, for the dual objectives of: (a) developing a test bed to validate volume reconstruction methods used to process clinical CMR data, via comparison with higher resolution data acquired using modalities available at the laboratory level, and (b) acquiring ventricular wall motion using CMR sequences for use as initial conditions in LV FSI models, in order to validate the predictive accuracy of the simulations through comparison with CMR-based flow field data
Summary
The development of clinically applicable fluid-structure interaction (FSI) models of the left heart is inherently challenging when using in vivo cardiovascular magnetic resonance (CMR) data for validation, due to the lack of a well-controlled system where detailed measurements of the ventricular wall motion and flow field are available a priori. Detailed studies examining the complex mechanical interactions between the various anatomical structures (left atrium, LV, aorta, and corresponding valves) on the pumping function are needed at the isolated structural levels and collective organ level. Such investigations can aid in improving clinical outcomes by identifying more accurate diagnostic measures for earlier intervention, as well as in optimizing treatment options a priori. The data obtained from these non-invasive techniques can be used in computational models [21, 23, 24, 28] in order to provide patient specific treatment options
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
