An Introductory Overview of Image-Based Computational Modeling in Personalized Cardiovascular Medicine

Figures

• 

  FIGURE 1. Process flow diagram outlining the image-based computational pipeline of the HeartFlow’s approach to calculating the fractional flow reserve computerized tomography (FFRCT). (Reproduced with permission from Nick Curzen, Professor of Interventional Cardiology/Consultant Cardiologist, University Hospital Southampton).

• 

  FIGURE 2. Macroscopic overview of the three physics occurring in the heart. (A) Cardiac conduction system schematic. (B) Blood flow path (navy blue arrows) and the fibrous structure defining the biomechanics of the heart wall (inset).

• 

  FIGURE 3. Schematic diagram of the heart tissue microstructure. (A) The transmural configuration of the muscle fibers and laminar sheets. (B) The layered organization of myocytes and the collagen fibers between the sheets. f0, fiber axis; s0, sheet axis and n0, sheet-normal axis. (Adopted with permission from Holzapfel and Ogden, 2009).

• 

  FIGURE 4. Comparison between (A) rule-based method and (B) DTI-based estimation of the myocardial fiber orientation for a 3D model of canine ventricles. (Reproduced with permission from Bayer et al., 2012).

• 

  FIGURE 5. Electrical coupling of the neighboring cardiomyocytes via the gap-junctions between their membranes.

• 

  FIGURE 6. Components of a multiscale cardiac electrophysiology model. (Left) Equations and sample output for a Hodgkin-Huxley formulation of the rapid sodium current through an ion channel. Multiple such sub-cellular models can be used to define a cell model. (Center) Schematic of sub-cellular processes included in a hypothetical cell model, together with the differential equation governing the transmembrane voltage, and sample output. Cell models differ in their formulation of the ionic current iion and can be made up of dozens of ordinary differential equations. (Right) Cell models can be incorporated into the bidomain equations and solved on a computational mesh of the heart (top right: high-resolution rabbit biventricular mesh of Bishop et al., 2010), to simulate normal or arrhythmic cardiac activity (bottom right). (Adopted with permission from Pathmanathan and Gray, 2018).

• 

  FIGURE 7. Summary of the Electrophysiology module’s inputs, outputs and applications. Superscripts in the figure correspond to the following references: 1Cardenes et al., 2014; Kayvanpour et al., 2015; Lopez-Perez et al., 2015; 2Cardenes et al., 2014; Palamara et al., 2014; 3Arevalo et al., 2016; Trayanova et al., 2017; 4Lopez-Perez et al., 2015; 5Behradfar et al., 2014; 6Arevalo et al., 2016; Deng et al., 2016; Trayanova and Chang, 2016; Prakosa et al., 2018; 7Boyle et al., 2016; Roney et al., 2018; 8Smith et al., 2011; Boyle et al., 2013; Potse et al., 2014; 9Trayanova, 2011; Behradfar et al., 2014; 10Arevalo et al., 2016; Trayanova et al., 2017; Deng et al., 2019.

• 

  FIGURE 8. (A) Sketch of the cardiac electro-fluid–structure coupling. (B) The three computational domains (fluid domain ΩF, solid mechanics domain ΩS, electrophysiology domain ΩE) considered in the cardiac multiphysics problem. In this example the electrophysiology physics are only considered in the ventricles, whereas the solid mechanics physics applies to the atria also. Additionally pictured are the fluid–structure interface ΓI inside the left ventricle, the epicardial surface ΓS,epi, and the mitral valve inlet surface Γin. Lastly, the domains ΩE and ΩS overlap within the ventricular part of the heart. (Adopted with permission from Quarteroni et al., 2017).

• 

  FIGURE 9. Sub-cellular tension generation mechanisms. (A) Organization of the cardiac muscle cell’s contraction mechanism (bottom inset shows the contraction-relaxation cycle of a single sarcomere). (B) Diagram showing how calcium release from internal cytosol storage causes cardiomyocyte contraction in response to the “outside-to-inside” calcium signaling.

• 

  FIGURE 10. Summary of the Biomechanics module’s inputs, outputs and applications. Superscripts in the figure correspond to the following references: 1Krishnamurthy et al., 2013; Meoli et al., 2015; Finsberg et al., 2018; 2Krishnamurthy et al., 2013; Meoli et al., 2015; 3Krishnamurthy et al., 2013; Kayvanpour et al., 2015; Meoli et al., 2015; Finsberg et al., 2018; Palit et al., 2018; Shavik et al., 2020; 4Krishnamurthy et al., 2013; 5Krishnamurthy et al., 2013; Palit et al., 2018; 6Lee et al., 2013; Meoli et al., 2015; Finsberg et al., 2018.

• 

  FIGURE 11. Two-element Windkessel circuit analogy illustrated.

• 

  FIGURE 12. Overview of the specific methods, modules and models commonly used in the state-of-the-art multiscale platelet hemodynamics and thrombus development studies. (Adopted with permission from Wang and King, 2012).