Accesso libero

Simulation studies on the hemodynamic model for blood flow

INFORMAZIONI SU QUESTO ARTICOLO

Cita

Introduction

Recently, in the field of biological rheology, rheology related to blood, vessels and heart, which constitutes human blood circulation, has been developed, rapidly [1]. The blood circulation is a complex system that affects the entire body. In order to better understand the physiological and pathological behavior of the cardiovascular system, it is important to study the dynamics of blood flow in the arteries and the mechanical factors of blood flow [2]. Therefore, studies concentrate on hemodynamics, which examines the properties of arterial blood flow movement. The characteristics of two structures are important here; blood and blood vessels/arteries [2, 3]. Blood is a special suspension of various blood cells and plasma, and its rheological properties are non-Newtonian fluid, but it is considered an incompressible inviscid fluid [4, 5]. Blood flow submits the principles of universal conservation of mass, momentum and energy [2, 6]. The forces that direct the blood flow are gravity and pressure gradient force, while the hindering forces are shear forces resulting from viscosity and turbulence [5,7]. The wall thickness to vessel radius (diameter) ratio due to arteries, are considered flat incompressible, thick viscoelastic [5, 8, 9] or elastic tubes [10], as well as thin elastic [5, 11,12,13] or viscoelastic tubes [5, 14, 15]. Therefore, these features show obvious nonlinearity. Usually, for simplicity, arteries are modeled as circular cylindrical elongated tubes with a constant cross-section [1, 5]. Under these assumptions, many models were proposed by well-known equations in the literature [3, 16,17,18,19]. Firstly, the Korteweg-de Vries (KdV) equation was determined by combining nonlinear science with hemodynamics for simplest assumptions [4, 20]. Later, considering more complex situations, the variable/constant coefficient forced KdV equation was determined [5, 21,22,23]. Boussinesq-type equations derived from the flow equations in elastic tubes and analyzed the effects of a local increase of radius followed by local variation of the thickness or rigidity of an elastic tube on the behavior of solitary waves [1, 4, 24]. The nonlinear Schrödinger equation was proposed in the view of Navier-Stokes equations and continuous equation [1, 4, 25].

In this work, the latest models will be considered, and the exact solutions are obtained in the soliton form and wave form. Most of the research in the literature is related to the model development, numerical and semi-analytical solutions. Therefore, we will try to fill the gap in this area by obtaining analytical solutions. To reach the aim, Bernoulli method, one of the ansatz-based methods, will be considered. The results will play an important role in supporting the improvement of models and the theoretical basis and diagnosis of some blood diseases.

Method summary

Ansatz methods first come to mind when searching for analytical solutions. Various types of ansatz methods are seen in the literature [26,27,28,29,30,31,32]. Generally, first step is the reduction, the classical wave transformation ξ = xct, c ≠ 0 is applied and then the model is reduced to nonlinear ordinary differential equation (NODE). Then, the assumption of solution is considered as follows; uξ=i=0Ngiziξ, u\left( \xi \right) = \sum\limits_{i = 0}^N {g_i}{z^i}\left( \xi \right), where gi (i = 0, ⋯ , N) are parameters that the methods aim is to determine them and N is defined by balancing principle [32]. Generally, z (ξ) is sometimes determined as a function and sometimes as a solution to an ordinary differential equation. In this work, z (ξ) is the solution of the Bernoulli differential equation (BDE) given as below zξ=Pzξ+Qz2ξ, z'\left( \xi \right) = Pz\left( \xi \right) + Q{z^2}\left( \xi \right), where P and Q parameters, and its solution is zξ=PC1PexpPξQ. z\left( \xi \right) = \frac{P}{{{C_1}P\exp \left( { - P\xi } \right) - Q}}. and C1 is an integration constant. At the last step, the assumption of the solution and if needed the BDE are substitute into NODE and by solving the system of equations obtained with coefficients of the powers of z (ξ), the mentioned parameters are obtained. Substituting all parameters and the solution of BDE into the solution assumption, the exact solution is extracted.

Applications

In this part of the paper, we apply the approach to obtain the exact solutions of governing models.

The blood flow and motion of arterial wall model

Considering that the blood is viscous fluid and the flow through a circular cylindrical tube of radius a. In the view of Navier-Stokes equation or equivalently the second law of Newton for fluid flow, the model is proposed [2]; utt(x,t)v02uxx(x,t)=12u(x,t)2tt+suxxtt(x,t)+βv02uxx(x,t), {u_{tt}}(x,t) - v_0^2{u_{xx}}(x,t) = \frac{1}{2}{\left( {u{{(x,t)}^2}} \right)_{tt}} + s\, {u_{xxtt}}(x,t) + \beta v_0^2{u_{xx}}(x,t), where v0=ηE2ρ1/2 {v_0} = {\left( {\frac{{\eta E}}{{2\rho }}} \right)^{1/2}} and s=ηρwR02ρ s = \frac{{\eta {\rho _w}{R_0}}}{{2\rho }} are determined. v0 is the phase velocity of nonlinear waves and depends on wall incompressibility (η), Young modulus (E) and density of blood (ρ), whereas sdepends on wall incompressibility (η), density of the artery wall (ρw), equilibrium for artery of radius (R0) and density of blood (ρ). β = α − 2 and α is the coefficient of nonlinear elasticity. As seen all parameters vary from individual to individual. Equation (1) is proposed for the nonlinear coefficient of elasticity of the artery wall and for dispersion connected to the inertial effects of the wall. Equation (1) is generally reduced to KdV equation via the reductive perturbation method and rescaling of coordinates [2, 33, 34] and the solutions obtained via known numerical or analytical methods [2]. Now, the classical wave transformation ξ = xvt, v ≠ 0 is used to reduce (1) to NODE and it is obtained v1+βv02uv2u2v2uusv2u(4)=0, \left( {v - \left( {1 + \beta } \right){v_0}^2} \right)u'' - {v^2}{\left( {u'} \right)^2} - {v^2}uu'' - s{v^2}{u^{(4)}} = 0, being u=dudξ u' = \frac{{du}}{{d\xi }} . Firstly, (2) is integrated twice respect to ξ and integration constants are assumed zero. One reads the equation (2) as below v1+βv02uv22u2sv2u=0. \left( {v - \left( {1 + \beta } \right){v_0}^2} \right)u - \frac{{{v^2}}}{2}{u^2} - s{v^2}u'' = 0.

For (3), the exact solution is assumed as follows uξ=g0+g1zξ+g2z2ξ, u\left( \xi \right) = {g_0} + {g_1}z\left( \xi \right) + {g_2}{z^2}\left( \xi \right), where g0, g1, g2 are parameters and z (ξ) is a solution of BDE z′ (ξ) = Pz (ξ) + Qz2 (ξ), P and Q are considered as constant functions. Therefore, substituting the assumptions in (3), a nonlinear algebraic equation system is obtained. The solution sets of the nonlinear algebraic system include trivial and constant solutions, the meaningful solution is given below; zξ=β+1α2+v2β+1α2+v2C1expτ+vQsα2β+1v2, z\left( \xi \right) = \frac{{ - \left( {\beta + 1} \right){\alpha ^2} + {v^2}}}{{\left( { - \left( {\beta + 1} \right){\alpha ^2} + {v^2}} \right){C_1}\exp \left( \tau \right) + vQ\sqrt {s\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right)} }}, where g0=2α2β+1v2v2 {g_0} = - \frac{{2\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right)}}{{{v^2}}} , g1=12α2β+1v2Qv2P {g_1} = - \frac{{12\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right)Q}}{{{v^2}P}} , g2=12α2β+1v2Q2v2P2 {g_2} = - \frac{{12\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right){Q^2}}}{{{v^2}{P^2}}} , P=±sα2β+1v2vs P = \pm \frac{{\sqrt {s\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right)} }}{{vs}} , α2β+1v20 \left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right) \ne 0 , τ=sα2β+1v2vsξ \tau = \frac{{\sqrt {s\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right)} }}{{vs}}\xi . When all parameters are substituted into the assumption of solution, the exact solution as the travelling wave is obtained. The 3D-plot of the exact solution for Q = 1, s = 2, v = 0.5, β = 3.55, α = 0.1, C1 = 0.1 is given in Figure 1. Additionally, for the same parameter’s values, 2D-plot is given at t = 0, t = 0.2, t = 0.4, t = 0.6, t = 0.8, respectively, in Figure 2.

Fig. 1

3D simulation of the exact solution of the (1) for Q = 1, s = 2, v = 0.5, β = 3.55, α = 0.1, C1 = 0.1

Fig. 2

2D simulation of the exact solution of the (1) for Q = 1, s = 2, v = 0.5, β = 3.55, α = 0.1, C1 = 0.1

The blood flow in the blood vessels

Considering that the blood is viscous fluid and the flow through a circular cylindrical tube of radius a. In the view of Navier-Stokes equation or equivalently the second law of Newton for fluid flow, the model is proposed [2]; utt(x,t)v02uxx(x,t)=12u(x,t)2tt+suxxtt(x,t)+βv02uxx(x,t), {u_{tt}}(x,t) - {v_0}^2{u_{xx}}(x,t) = \frac{1}{2}{\left( {u{{(x,t)}^2}} \right)_{tt}} + s {u_{xxtt}}(x,t) + \beta {v_0}^2{u_{xx}}(x,t), being v0=ηE2ρ1/2 {v_0} = {\left( {\frac{{\eta E}}{{2\rho }}} \right)^{1/2}} and s=ηρwR02ρ s = \frac{{\eta {\rho _w}{R_0}}}{{2\rho }} are determined. v0 is the phase velocity of nonlinear waves and depends on wall incompressibility (η), Young modulus (E) and density of blood (ρ), whereas sdepends on wall incompressibility (η), density of the artery wall (ρw), equilibrium for artery of radius (R0) and density of blood (ρ). β = α − 2 and α is the coefficient of nonlinear elasticity. As seen all parameters vary from individual to individual. (1) is proposed for the nonlinear coefficient of elasticity of the artery wall and for dispersion connected to the inertial effects of the wall. (1) is generally reduced to KdV equation via the reductive perturbation method and rescaling of coordinates [2,33,34] and the solutions obtained via known numerical or analytical methods [2]. Now, the classical wave transformation ξ = xvt, v ≠ 0 is used to reduce (1) to NODE and it is found v1+βv02uv2u2v2uusv2u(4)=0, \left( {v - \left( {1 + \beta } \right){v_0}^2} \right)u'' - {v^2}{\left( {u'} \right)^2} - {v^2}uu'' - s{v^2}{u^{(4)}} = 0, where u=dudξ u' = \frac{{du}}{{d\xi }} . Firstly, (2) is integrated twice respect to ξ and integration constants are assumed zero v1+βv02uv22u2sv2u=0. \left( {v - \left( {1 + \beta } \right){v_0}^2} \right)u - \frac{{{v^2}}}{2}{u^2} - s{v^2}u'' = 0.

For (3), the exact solution is assumed as follows uξ=g0+g1zξ+g2z2ξ, u\left( \xi \right) = {g_0} + {g_1}z\left( \xi \right) + {g_2}{z^2}\left( \xi \right), where g0, g1, g2 are parameters and z (ξ) is a solution of BDE z′ (ξ) = Pz (ξ) + Qz2 (ξ), P and Q are considered as constant functions. Therefore, substituting the assumptions in (3), a nonlinear algebraic equation system is obtained. The solution sets of the nonlinear algebraic system include trivial and constant solutions, the meaningful solution is given below. zξ=β+1α2+v2β+1α2+v2C1expχ+vQsα2β+1v2, z\left( \xi \right) = \frac{{ - \left( {\beta + 1} \right){\alpha ^2} + {v^2}}}{{\left( { - \left( {\beta + 1} \right){\alpha ^2} + {v^2}} \right){C_1}\exp \left( \chi \right) + vQ\sqrt {s\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right)} }}, where g0=2α2β+1v2v2 {g_0} = - \frac{{2\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right)}}{{{v^2}}} , g1=12α2β+1v2Qv2P {g_1} = - \frac{{12\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right)Q}}{{{v^2}P}} , g2=12α2β+1v2Q2v2P2 {g_2} = - \frac{{12\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right){Q^2}}}{{{v^2}{P^2}}} , P=±sα2β+1v2vs P = \pm \frac{{\sqrt {s\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right)} }}{{vs}} , α2β+1v20 \left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right) \ne 0 , χ=sα2β+1v2vsξ \chi = \frac{{\sqrt {s\left( {{\alpha ^2}\left( {\beta + 1} \right) - {v^2}} \right)} }}{{vs}}\xi . When all parameters are substituted into the assumption of solution, the exact solution as the travelling wave is obtained. The 3D-plot of the exact solution for Q = 1, s = 2, v = 0.5, β = 3.55, α = 0.1, C1 = 0.1 is given in Figure 3. Additionally, for the same parameter’s values, 2D-plot is given at t = 0, t = 0.2, t = 0.4, t = 0.6, t = 0.8, respectively, in Figure 4.

Fig. 3

3D simulation of the exact solution of the (1) for Q = 1, s = 2, v = 0.5, β = 3.55, α = 0.1, C1 = 0.1.

Fig. 4

2D simulation of the exact solution of the (1) for Q = 1, s = 2, v = 0.5, β = 3.55, α = 0.1, C1 = 0.1.

Conclusions

The aim of the work is to obtain wave solutions of hemodynamic models for blood flow that is considered motion of arterial wall and in blood vessels, respectively. In the literature, most of the research is depending on the improvement of model whereas some are depended on to obtain the numerical and semi-analytical solutions. Therefore, we tried to fill the gap in this area that the solutions were obtained analytically and 3D- and 2D-plots were proposed. The solutions were proposed in the explicit form and the parameters were considered arbitrarily for healthy human, so this parameter values are depended on individual health. If human has a health problem, we can say that the wave structure will change due to the individual parameters. Therefore, the results will support the theoretical basis and in diagnosis for some blood diseases. Additionally, the results will be important role to improve the models. For future works, it is interested in applying other well-organized methods, such as Bifurcation analysis [30], Jacobi elliptic function method [29], Kudryashov method [27], etc. to derive other wave solutions of the governing models.

Declarations
Conflict of interest 

The authors have no competing interests to declare that are relevant to the content of this article.

Funding

No funding was received to assist with the preparation of this manuscript.

Author’s contribution

Z.P.I.-Conceptualization, Methodology, Software, Writing-Review Editing, Formal Analysis, Validation, Writing-Original Draft. The author read and approved the final submitted version of this manuscript.

Acknowledgement

The author deeply appreciate the anonymous reviewers for their helpful and constructive suggestions, which can help further improve this paper.

Data availability statement

All data that support the findings of this study are included within the article.

Using of AI tools

The authors declare that they have not used Artificial Intelligence (AI) tools in the creation of this article.

eISSN:
2956-7068
Lingua:
Inglese
Frequenza di pubblicazione:
2 volte all'anno
Argomenti della rivista:
Computer Sciences, other, Engineering, Introductions and Overviews, Mathematics, General Mathematics, Physics