Towards non-invasive computational-mechanics and imaging-based diagnostic framework for personalized cardiology for coarctation

Abstract

coarctation of the aorta (coA) is a congenital narrowing of the proximal descending aorta. Although accurate and early diagnosis of COA hinges on blood flow quantification, proper diagnostic methods for COA are still lacking because fluid-dynamics methods that can be used for accurate flow quantification are not well developed yet. Most importantly, coA and the heart interact with each other and because the heart resides in a complex vascular network that imposes boundary conditions on its function, accurate diagnosis relies on quantifications of the global hemodynamics (heart-function metrics) as well as the local hemodynamics (detailed information of the blood flow dynamics in COA). In this study, to enable the development of new non-invasive methods that can quantify local and global hemodynamics for coA diagnosis, we developed an innovative fast computational-mechanics and imaging-based framework that uses Lattice Boltzmann method and lumped-parameter modeling that only need routine non-invasive clinical patient data. We used clinical data of patients with coA to validate the proposed framework and to demonstrate its abilities to provide new diagnostic analyses not possible with conventional diagnostic methods. We validated this framework against clinical cardiac catheterization data, calculations using the conventional finite-volume method and clinical Doppler echocardiographic measurements. the diagnostic information, that the framework can provide, is vitally needed to improve clinical outcomes, to assess patient risk and to plan treatment. Coarctation of the aorta (COA) is a congenital narrowing of the proximal descending aorta. The hemodynamic severity and clinical manifestations of COA vary from asymptomatic mild narrowing of the aortic isthmus to severe obstruction associated with cardiac defects, congestive heart failure and shock in the neonatal period, persistent hypertension and aortic dissection 1. Not all patients are symptomatic but with disease progression in severity, 60% of adults over 40 with uncorrected COA develop heart failure and 75% of them die by the age of 50, and 90% of them die by the age of 60 2. Indeed, despite advancements in interventional/surgical techniques, the long-term morbidity and subsequent mortality of patients with COA remain high in comparison with the general population 3,4. "Cardiology is flow" 5 and therefore the essential sources of COA morbidity can be explained on the basis of adverse hemodynamics: abnormal biomechanical forces, abnormal flow patterns-that often characterized by disturbed and turbulent flow-and in some cases by an increase in the heart workload that leads to the development and progression of cardiovascular diseases 5-8. Flow quantification can be greatly useful for accurate and early diagnosis, but we still lack proper diagnostic methods for many cardiovascular diseases 6,9 , including COA, because the fluid-dynamics methods that can be used as engines of new diagnostic tools are not well developed yet. In this research we contributed to advanc...

Comparison of LBM and FVM in the estimation of LAD stenosis

“…Therefore, as an alternative to FVM, LBM is also applied for haemodynamic simulations. [14][15][16] Previous comparison of LBM and FVM in the haemodynamic simulation of a bifurcation in abdominal arteries confirmed that LBM is able to deliver comparable accurate results with less computational cost. Furthermore, the lattice generation process is much easier than FVM grid generation.…”

Effects of Choice of Medical Imaging Modalities on a Non-invasive Diagnostic and Monitoring Computational Framework for Patients With Complex Valvular, Vascular, and Ventricular Diseases Who Undergo Transcatheter Aortic Valve Replacement

“…In the second step, R SA , C SAC , and C ao were optimized so that maximum and minimum of the aorta pressures were equal to the systolic pressure and diastolic pressure, respectively, measured using a sphygmomanometer in each patient. Because the left ventricle confronts the total systemic resistance and not the specific resistances, and the systemic arteries resistance ( R SA ) is one order of magnitude greater than both the aortic resistance ( R ao ) and systemic vein resistance ( R SV ), we considered R ao and R SV as constants and optimized R SA as the main contributor of the total systemic resistance ( Keshavarz-Motamed et al, 2011 , 2012 , 2014 , 2015 ; Benevento et al, 2015 ; Keshavarz-Motamed, 2020 ; Sadeghi et al, 2020 ). C ao was considered to be 0.6 of C SAC because 60% of the total arterial compliance lives in the proximal aorta ( Stergiopulos et al, 1999 ).…”

“…“Cardiology is flow” ( Richter and Edelman, 2006 ) and indeed quantifications of hemodynamics can be immensely valuable for precise diagnosis, however, we still lack precise diagnostic tools for various cardiovascular diseases ( Di Carli et al, 2016 ). There has been an emerging conclusion by many researchers that valvular disease is a complex and mixed disease that also depends on the ventricle and the vascular system states ( Yin, 1987 ; Burkhoff et al, 2005 ; Borlaug and Kass, 2008 ; Taylor and Steinman, 2010 ; Dweck et al, 2012 ; Antonini-Canterin et al, 2013 ; Keshavarz-Motamed et al, 2014 , 2016 ; Ben-Assa et al, 2019 ; Ikonomidis et al, 2019 ; Keshavarz-Motamed et al, 2020 ; Sadeghi et al, 2020 ; Khodaei et al, 2021a , b ). Indeed, the quantitative investigations of hemodynamics in patients with C3VI should take into account the interactive coupling of the valves, ventricle, and the vascular system.…”

“…The conclusions and recommendations made in the previous studies can be boiled down to define the following two hemodynamics quantification capabilities that computational diagnostic frameworks are required to have to be clinically useful for patients with C3VI. The required quantities are local and global hemodynamics metrics ( Yin, 1987 ; Burkhoff et al, 2005 ; Borlaug and Kass, 2008 ; Taylor and Steinman, 2010 ; Dweck et al, 2012 ; Antonini-Canterin et al, 2013 ; Ky et al, 2013 ; Keshavarz-Motamed et al, 2014 , 2016 ; Ben-Assa et al, 2019 ; Ikonomidis et al, 2019 ; Seemann et al, 2019 ; Keshavarz-Motamed et al, 2020 ; Sadeghi et al, 2020 ) as follows: (1) Metrics of circulatory function (local) , e.g., fluid dynamics of the circulatory system, and (2) Metrics of cardiac function (global) , e.g., heart workload and its breakdown to the contributing disease components. Assessments of hemodynamics, if available, would offer valued information about the cardiac health condition and could be used for planning C3VI interventions and making critical clinical decisions with life-threatening risks.…”

The Critical Role of Lumped Parameter Models in Patient-Specific Cardiovascular Simulations

“…"Cardiology is flow" [14] and indeed hemodynamic quantification can be immensely valuable for precise and early diagnosis, however, we still lack precise and early diagnostic tools for various cardiovascular diseases because the hemodynamics analysis methods that can be used as engines of such diagnostic tools are not well developed yet [10]. The required quantities for these tools are local and global hemodynamics metrics [15][16][17][18][19][20][21][22][23][24][25][26][27][28]: (1) Metrics of circulatory function (local), e.g., detailed information of the dynamics of the circulatory system, and (2) Metrics of cardiac function (global), e.g., heart workload and its contribution breakdown of each component of the cardiovascular diseases. Assessments of hemodynamics, if available, would provide valuable information about the patient's state of cardiac deterioration as well as heart recovery and could be used for planning complex valvular-vascular-ventricular disease (C3VD) interventions and making critical clinical decisions with life-threatening risks.…”