SIMULATION OF BLOOD FLOW THROUGH

MECHANICAL HEART VALVE USING ANSYS FLUID

STRUCTURE INTERACTION

By:

CHAI JIEN WEI

(Matrix no: 120363)

Supervisor:

Dr. Mohamad Aizat Abas

JUNE 2017

This dissertation is submitted to

Universiti Sains Malaysia
As partial fulfillment of the requirement to graduate with honors degree in
BACHELOR OF ENGINEERING (MECHANICAL ENGINEERING)

School of Mechanical Engineering

Engineering Campus
Universiti Sains Malaysia
 Declaration

This work has not previously been accepted in substance for any degree and is not being

concurrently submitted in candidature for any degree.

Signed.................................................................................... (CHAI JIEN WEI)

Date.......................................................................................

STATEMENT 1

This thesis is the result of my own investigations, except where otherwise stated. Other

sources are acknowledged by giving explicit references. Bibliography/references are

appended.

Signed.................................................................................... (CHAI JIEN WEI)

Date........................................................................................

STATEMENT 2

I hereby give consent for my thesis, if accepted, to be available for photocopying and for

interlibrary loan, and for the title and summary to be made available outside organizations.

Signed..................................................................................... (CHAI JIEN WEI)

Date.........................................................................................

i
 Acknowledgement
First and foremost, I would like to express my deepest appreciation to my
supervisor Dr. Mohamad Aizat Abas for his continuous supervision, assistance and support
from the initial to the final stage of my final project. This dissertation would not be possible
without his patient guidance, enthusiastic encouragement and constructive suggestions
throughout the whole period. The valuable lessons learnt not only helped me in enhancing
my knowledge, but the interpersonal skill and the positive attitude as well.

Besides, I would also like to thank to master student, kak Hafizah for providing me
useful advices and assistance when doing experiment. Sincere thanks are also extended to
all the technicians for the given technical support. They are keen in teaching and assisting
me when dealing with the machines and various equipment during my laboratory work.
Their careful and precious guidance were extremely valuable for my study both
theoretically and practically.

Next I would like to express my sincere gratitude to my family who constantly

support and encourage me spiritually and mentally. A big thanks to my fellow coursemates
for sharing their knowledge and ideas in the construction of this project.

Last but not least, I wish to express my sincere thanks to all the lecturers involved
in my final year project for their precious guidance and opinions which are extremely
valuable for my study both theoretically and practically. I perceive this research project as
a big milestone in my career development. I will strive to use gained skills and knowledge
in the best possible way, and I will continue to work on their improvement, in order to
attain desired career objectives.

ii
 Table of Contents
Declaration ........................................................................................................................... i

Acknowledgement .............................................................................................................. ii

Table of Contents ............................................................................................................... iii

List of Tables ..................................................................................................................... vi

List of Figures ................................................................................................................... vii

List of Abbreviation ........................................................................................................... ix

Abstrak ................................................................................................................................ x

Abstract .............................................................................................................................. xi

: Introduction ...................................................................................................... 1

1.1 Research Background ............................................................................................... 1

1.2 Problem Statement .................................................................................................... 3

1.3 Objectives ................................................................................................................. 3

1.4 Scope of Project ........................................................................................................ 3

: Literature Review ............................................................................................. 4

: Methodology .................................................................................................... 9

3.1 Simulation using ANSYS Fluid Structure Interaction (FSI) .................................... 9

3.1.1 Pre-Analysis ..................................................................................................... 10

3.1.1.1 Governing Equations ................................................................................ 10

3.1.1.2 Boundary Conditions ................................................................................ 11

3.1.2 Establishment of CAD Model .......................................................................... 13

3.1.3 Transient Structural .......................................................................................... 15

3.1.3.1 Geometry................................................................................................... 16

3.1.3.2 Mesh .......................................................................................................... 17

3.1.3.3 Physics Setup ............................................................................................ 18

iii
 3.1.4 Fluid Flow (FLUENT) ..................................................................................... 19

3.1.4.1 Mesh .......................................................................................................... 19

3.1.4.2 Physics Setup ............................................................................................ 21

3.1.5 System Coupling .............................................................................................. 23

3.2 Experiment using Particle Image Velocimetry (PIV) ............................................. 25

3.2.1 Fabrication of Experimental Model ................................................................. 25

3.2.2 Working Fluid .................................................................................................. 26

3.2.3 Experiment Setup ............................................................................................. 27

3.2.4 PIVlab .............................................................................................................. 29

: Results and Discussion ................................................................................... 30

4.1 Verification of Simulation Model ........................................................................... 30

4.1.1 Mass Conservation ........................................................................................... 30

4.1.2 Inlet Boundary Condition ................................................................................ 31

4.1.3 Mesh Refinement ............................................................................................. 31

4.1.4 Smaller Time Step............................................................................................ 32

4.1.5 Experiment Validation ..................................................................................... 33

4.2 Velocity Contour ..................................................................................................... 35

4.3 Doppler Velocity Index (DVI) ................................................................................ 36

4.4 Vorticity .................................................................................................................. 40

4.5 Pressure Contour ..................................................................................................... 41

4.6 Wall Shear............................................................................................................... 43

4.7 Von Mises Stress and Strain ................................................................................... 45

4.8 Selection of the Best Leaflets Design ..................................................................... 48

: Conclusions .................................................................................................... 50

5.1 Future Work ............................................................................................................ 51

iv
References ......................................................................................................................... 52

Appendices ........................................................................................................................ 55

Appendix 1: UDF of Inlet Velocity .............................................................................. 55

Appendix 2: Flow sensor circuit connection ................................................................ 56

Appendix 3: 2D drawing of CAD model ...................................................................... 61

v
 List of Tables
Table 4.1. Maximum pressure at left atrioventricular wall for different relevance .......... 31
Table 4.2. Coordinates of the selected ten points for data extraction ............................... 36
Table 4.3. Average velocity for different leaflets curvature ............................................. 37
Table 4.4. DVI of different leaflets curvature for both simulation and experiment ......... 39
Table 4.5. Maximum wall shear stress for different leaflets curvature ............................ 45
Table 4.6. Maximum Von Mises stress and safety factor for different leaflets curvature 47
Table 4.7. Ranking of the design parameters for different leaflets curvature................... 48

vi
 List of Figures
Figure 1.1. Four chambers and four valves of human heart .............................................. 1
Figure 1.2. Thrombosis on mitral valve prosthesis entirely covering inflow aspect of one
flap ...................................................................................................................................... 2
Figure 1.3 Mitral valve vegetation ..................................................................................... 2
Figure 1.4. On-X prosthetic valve structure ....................................................................... 3
Figure 2.1. Schematic diagram of steady flow loop setup .................................................. 7
Figure 2.2. Schematic overview of the experimental setup ................................................ 8
Figure 3.1. General 2-way FSI setup procedure ................................................................. 9
Figure 3.2. Inlet velocity profile in two periods .............................................................. 12
Figure 3.3. Simplified CAD model for simulation with predefined dimensions .............. 14
Figure 3.4. Explanation of leaflets curvature modifications ............................................. 14
Figure 3.5. Different leaflets curvature of the mechanical heart valve. ............................ 15
Figure 3.6. General physical properties of pyrolytic carbon ............................................ 15
Figure 3.7. Physical properties of pyrolytic carbon that used in ANSYS simulation ...... 16
Figure 3.8. Geometry imported to the ANSYS Design Modeler ...................................... 16
Figure 3.9. Four edges that are highlighted for refinement .............................................. 17
Figure 3.10. The meshed geometry(leaflets) .................................................................... 17
Figure 3.11. Face that is selected for the fixed to the ground joint................................... 18
Figure 3.12. Faces that are chosen for the revolute joints ................................................ 18
Figure 3.13. 18 faces that are selected for Fluid Solid Interface ...................................... 19
Figure 3.14. Geometry before and after apply virtual topology ....................................... 19
Figure 3.15. Faces selected for named selection............................................................... 20
Figure 3.16. Range of skewness and orthogonal quality to gauge the quality of mesh ... 21
Figure 3.17. Cross-sectional view of the meshing of the model ....................................... 21
Figure 3.18. Parameters for smoothing(left) and remeshing(right) mesh method ............ 22
Figure 3.19. Mesh distribution of the left heart model .................................................... 23
Figure 3.20. Flow chart showing how the system coupling works in ANSYS ................ 24
Figure 3.21. Two way FSI setup in ANSYS ..................................................................... 24
Figure 3.22. Parameters for the data transfers in system coupling interface .................... 25
Figure 3.23. Rapid prototyped bileaflet heart valve ......................................................... 26

vii
Figure 3.24. Experimental model for PIV measurement .................................................. 26
Figure 3.25. Schematic diagrams of the flow system ....................................................... 27
Figure 3.26. Instantaneous flow rate of the experiment.................................................... 28
Figure 3.27. Control system in DynamicStudio................................................................ 28
Figure 3.28. PIV experimental setup ................................................................................ 29
Figure 3.29. Region of interest and masks applied using PIVlab ..................................... 29
Figure 4.1. Mass flux report of inlet and outlet ................................................................ 30
Figure 4.2. Velocity profile at inlet ................................................................................... 31
Figure 4.3. Maximum pressure at left atrioventricular wall over different number of
elements ............................................................................................................................ 32
Figure 4.4. Pressure at the left atrioventricular wall when the leaflets are fully opened. . 33
Figure 4.5. Comparison of velocity contour between simulation and experiment ........... 34
Figure 4.6. Comparison of vorticity between simulation and experiment ........................ 34
Figure 4.7. Velocity contour in x-y plane when the valves are fully opened. .................. 35
Figure 4.8. Selected ten points(A-J) for velocity data extraction ..................................... 36
Figure 4.9. Average Velocity for different leaflets curvature at the inlet ......................... 37
Figure 4.10. Average Velocity for different leaflets curvature at the outlet ..................... 38
Figure 4.11. Position of VLVO and Vjet in a prosthetic mitral valve .................................. 39
Figure 4.12. Doppler velocity index over leaflets curvature ............................................ 39
Figure 4.13. Doppler velocity index for a normal prosthetic mitral valve ....................... 40
Figure 4.14. Vorticity study in the x-y plane for different leaflets curvature. .................. 41
Figure 4.15. Development of pressure contour over a cardiac cycle. ............................... 43
Figure 4.16. Wall shear stress distribution for different leaflets curvature. ..................... 44
Figure 4.17. Von Mises stress over a cardiac cycle for different leaflets curvature ......... 46
Figure 4.18. Von Mises strain over a cardiac cycle for different leaflets curvature ......... 46
Figure 4.19. Maximum Von Mises stress at the hinge area .............................................. 48
Figure 4.20. Safety factor of the heart valve ..................................................................... 48
Figure 4.21. Finalized mechanical heart valve design. ..................................................... 49

viii
 List of Abbreviation

FSI Fluid Structure Interaction/ Fluid Solid Interface

PIV Particle Image Velocimetry

MHS Mechanical Heart Valve

ATS A company’s name

AHV Artificial Heart Valve

CFD Computational Fluid Dynamics

PHV Prosthetic Heart Valve

BMHV Bileaflet Mechanical Heart Valve

Re Reynold’s Number

CAD Computer Aided Design

UDF User Defined Function

FEA Finite Element Analysis

CNC Computer Numerical Control

STL STereoLithography

DVI Doppler Velocity Index

LVO Left Ventricular Outflow

ix
 Abstrak

Katup bileaflet merupakan katup jantung buatan yang unggul dan telah diimplan
buat kali pertama sejak daripada 20 tahun yang lalu. Ciri signifikan katup bileaflet ialah
geometri dua leaflet yang memainkan peranan penting dalam menentukan medan aliran.
Tujuan penyelidikan ini adalah untuk memantau corak aliran darah semasa darah melepasi
katup jantung bagi kelengkungan leaflets yang berlainan. Tambahan pula, analisis struktur
leaflets turut dikaji. Simulasi dilaksanakan menggunakan ANSYS Fluid Structure
Interaction (FSI). Experimen Particle Image Velocimetry jua akan dijalankan bagi
mengesahkan keputusan simulasi. Akhirnya, keputusan seperti kontur halaju dan
kepusaran dibandingkan dan mendapati bahawa gerakan leaflet dan medan aliran antara
simulasi numerikal dengan experimen mempunyai persamaan yang agak banyak. Dalam
kajian ini, leaflets yang melengkung ke arah dalam dengan sudut yang besar merupakan
configurasi terbaik disebabkan dinamik aliran darah yang lancar. Di samping itu, tekanan
von mises adalah paling tinggi di bahagian engsel. Semua keputusan akan berfungsi
sebagai asas dalam usaha mencipta katup jantung yang dapat meningkatkan hemodinamik
jantung.

x
 Abstract

A successful mechanical prosthetic heart valve design is the bileaflet valve, which
has been implanted for the first time more than 20 years ago. A key feature of bileaflet
valves is the geometry of the two leaflets, which can be very important in determining the
flow field. The aim of this research is to observe the flow pattern of the blood through the
heart valve with different leaflets curvatures. In addition, the structural analysis of the
leaflets was investigated. The simulation is carried out using ANSYS Fluid Structural
Interaction(FSI). Particle image velocimetry experiment will also be conducted to validate
the simulation results. Finally, the results such as velocity contour and vorticity are
compared, revealing great similarity in leaflet motion and flow fields between the
numerical simulation and the experimental test. Also, in this study, the leaflets which
curved inwards with greater degrees are found to be the best configuration as it allows the
greatest blood flow dynamic. Moreover, the maximum von mises stress is found to be at
the hinge region. These results will serve as a basis for valve design to improve the
hemodynamic of the heart.

xi
 : Introduction
The heart is a vital muscular organ which pumps blood throughout the body. Heart
valves are passive tissues of the heart which react to inertial forces exerted by surrounding
blood by opening and closing action. When functioning properly, heart valves as fluidic
control components of the heart will ensure unidirectional blood flow during the cardiac
cycle, with highest flow rate and lowest flow resistance. The four valves of the human heart
as shown in Figure 1.1 are the aortic, mitral, pulmonary and tricuspid valves. In this project,
mitral valve will be the prime focus.

Figure 1.1. Four chambers and four valves of human heart [1]

1.1 Research Background

Valvular stensosis, valvular regurgitation and valvular atresia are valvular heart
disease that are life-threatening and afflicts millions of people worldwide. Approximately
250,000 valves are repaired or replaced each year.[2] Malfunction of a native valve impairs
its efficient fluid mechanic and hemodynamic performance. Artificial heart valves have
been used since 1960 to replace diseased native valves and have saved millions of lives.
Unfortunately, despite five decades of use, these devices are less than ideal and lead to
many complications such as thrombosis, vegetation and hemolysis.

Thrombosis is the formation of a blood clot inside a blood vessel, obstructing the
flow of blood through the circulatory system. Thrombosis of a prosthetic valve is
potentially life threatening, resulting in hemodynamically severe stenosis or regurgitation.
According to Vuyisile T. Nkomo, M.D., M.P.H., director of the Valvular Heart Disease
Clinic at Mayo Clinic in Rochester, Minnesota, thrombotic risk is related to the type of
valve, position of the valve and adequacy of anticoagulation.[3]
1
Figure 1.2. Thrombosis on mitral valve prosthesis entirely covering inflow aspect of one flap [4]

In medicine, vegetation is often associated with endocarditis. Endocarditis is an

inflammation of the valves of the heart and it is often caused by the growth of bacteria on
one of the heart valves, leading to a mass known as a vegetation.[5] The big lump of mass
will bring serious complications as it impedes the flow of blood through the circulatory
system.

Figure 1.3 Mitral valve vegetation [6]

Hemolysis which is the rupturing of red blood cells is one of the potentially serious
complications of prosthetic heart valves. It is usually associated with either structural
deterioration or paravalvular leak. Therefore, the damage threshold of blood components
by fluid shear stress must be considered when designing a prosthetic heart valve.

Generally, a mechanical heart valve consists of a ring outer body, inner ring, leaflets,
orifice and hinges. Most mechanical heart valves are made of titanium, graphite, pyrolytic
carbon, and polyester. The titanium is used for the housing or outer ring, graphite coated
with pyrolytic carbon is used for the leaflet and 100% pyrolytic carbon is used for the inner
ring. Figure 1.4 shows one of the existing mechanical heart valve in the market which is
the On-X prosthetic bileaflet valve.

2
 Figure 1.4. On-X prosthetic valve structure [7]

1.2 Problem Statement

Today, many people live long and healthy lives and never realize they have a mild
valve problem. However, valve disease can seriously increase a persons’ risk for sudden
death or cause rapid development of problems in and around the heart that can become
fatal without treatment. The creation of mechanical heart valve has made an incredible
impact in biomedical field due to its lifetime durability. However, there are still problems
in mechanical valves such as thrombosis, vegetation, and many more. Hence, a better valve
design has to be developed to maximize the survival rate of patients.

1.3 Objectives
 To design a mechanical heart valve that meet forward and backward flow shear,
reduce thrombosis, hemolysis and cavitation.
 To validate the simulation results by conducting in vitro experiment

1.4 Scope of Project

Basically, the scope of work includes the design phase, simulation phase and
experiment. For the design phase, left heart model with five different leaflets
configurations are established using SolidWorks. Next, the created geometries are used in
the subsequent simulation phase. The simulation is carried out using ANSYS Fluid
Structure Interaction (FSI) since multiphysics problem is involved. Finally, PIV
experiment is conducted to compare the results with the simulation for validation purpose.
The left heart model used in both simulation and experiment are similar to the actual heart
of human to ensure that the results obtained are applicable to the real world.

3
 : Literature Review
History of Mechanical Heart Valve

The past 50 years have witnessed remarkable progress in the development of safe,
hemodynamically favorable mechanical heart valves. The need for prosthetic heart valves
was long recognized but seemed an impossible dream before 1952 when Dr Charles
Hufnagel clinically introduced a ball valve that he placed into the descending thoracic aorta
for treatment of aortic valvular insufficiency. This valve was implanted in a 30-year-old
woman who could lead a normal life after the surgery. [8]

According to Dr Charles Hufnagel research, the drawback of this design was that it
could only be placed in the descending aorta instead of the heart itself. Consequently, it did
not fully correct the valve problem but only alleviate the symptoms. However, it was a
significant progress because it proved that synthetic materials could be used to create heart
valves. [9]

In 1960, a new type of valve was successfully implanted which is the Starr-Edwards
ball valve. This valve was a modification of Hufnagel’s original valve. [10] The ball of the
valve was slightly smaller and caged from both sides so it could be inserted into the heart
itself.

Tilting discs were introduced in the later 1960s. These valves were a major
improvement over the ball designs as they allowed blood to flow in a more natural way
while reducing damage to blood cells from mechanical forces. Regrettably, the struts of
these valves tended to fracture from fatigue over time. [11]

Bileaflet valves were introduced in 1979. Blood flows directly through the center
of these valves (like in an intact heart valve) which makes these valves superior to other
designs. The major downside of this design is that it allows some backflow.

According to Journal of the Practice of Cardiovascular Sciences [12], the

thrombogenicity in caged ball is the highest followed by tilting disc and bileaflet and this
is why a vast majority of mechanical heart valves used today have bileaflet design.

4
Design Optimization of Mechanical Heart Valve

The paper by Yared Alemu [13] has previously compared two MHV, ATS and the
St. Jude Medical, and demonstrated that owing to its non-recessed hinge design, the ATS
valve offers improved thrombogenic performance. In the study, they further optimize the
ATS valve thrombogenic performance by modifying various design features of the valve,
intended to achieve reduced thrombogenicity. The modifications include optimizing the
leaflet-housing gap clearance, increasing the effective maximum opening angle of the
valve and introducing a streamlined channel between the leaflet stops of the valve that
increases the effective flow area.

Analysis of Flow Field in Mechanical Heart Valves Using Finite Volume Method

In the paper by Feng Zhou [14], the impact of valve structure on blood flow in aorta,
including hemodynamic aspects are investigated. The regional distribution of flow shear
stress in an artificial heart valve (AHV) is analyzed using computational fluid dynamics
and the AHV performance is evaluated in terms of variation of flow velocity and pressure
when blood passes the leaflets in the aortic valve. The findings indicate that for the design
of a mechanical AHV, the maximum opening angle and internal orifice diameter should be
increased to improve the blood flow dynamics and reduce the likelihood of damage to
blood components.

Besides, Mushtak Al-Atabi [15] investigated on blood flow through the mitral valve.
The study is done numerically using computational fluid dynamics. A two-dimensional,
2D model of the experimental design was simulated using ANSYS FLUENT. From the
results, it is seen that the flow through an open mitral valve produces two vortices. Vortices
are vital as they help propel the flow from the ventricle. The absence and presence of the
vortices and the structure of the vortices can be used to identify possible mitral valve
anomalies. This can be used as the baseline for the protocol.

Computational Modelling of Bileaflet Mechanical Valves Using Fluid-Structure

Interaction Approach

The study by Han Hung Yeh [16] presents numerical simulations of a minimally
constrained mechanical valve model using a fully coupled fluid-structure interaction

5
method with COMSOL Multiphysics. The model employs a physiological pulsatile
pressure gradient across an aortic valve with an approximately symmetric aortic root. The
complex hinge from the exact model is simplified with a pin joint and weak constraints to
control the designated valve leaflet positions. Also, arbitrary Lagrangian-Eulerian method
is exercised in order to accommodate large mesh displacements due to leaflet motion. It is
found that vortices were generated with higher blood velocity passing through the
unconstrained leaflet, which may lead to diagnostic confusion.

PIV Measurements of Flows in Artificial Heart Valves

The proposed study by Costantino Del Gaudio [17] presents the fabrication and in
vitro characterization of a biodegradable electrospun heart valve prosthesis using the
particle image velocimetry technique either in physiological and pathological fluid
dynamic conditions. The scaffold was designed to reproduce the aortic valve geometry,
also mimicking the fibrous structure of the natural extracellular matrix. To evaluate its
performances for possible implantation, the flow fields downstream the valve were
accurately investigated and compared. The experimental results showed a correct
functionality of the device, supported by the formation of vortex structures at the edge of
the three cusps, with Reynolds stress values below the threshold for the risk of hemolysis
(which can be comprised in the range 400–4000 N/m2 depending on the exposure period),
and a good structural resistance to the mechanical loads generated by the driving pressure
difference.

Rodoslav Kaminsky [18] proposed various modifications of PIV in order to

describe, compare and realize which method is the most suitable for the quantification of
such flows. The choice of the experimental procedure for testing the PHVs is strongly
dependent on the optical access of the designed in-vitro testing loops simulating the human
heart and vascular system. The hardware demand and its configuration such as stereoscopic
PIV is much more complex than standard 2D PIV, thus the conditions and design of the
testing loop have to be aware of to allow the desired flow measurement. The flow in heart
valves as an unsteady periodically generated flow, can be attained by averaged phase
locked or measurements with high temporal.

6
 Moreover, Majid Y. Yousif [19] revealed that for accurate particle image
velocimetry measurements in hemodynamics studies it is important to use a fluid with a
refractive index (n) matching that of the vascular models (phantoms) and ideally a dynamic
viscosity matching human blood.

Experimentally Validated Hemodynamics Simulations of Mechanical Heart Valves

The paper by Vinh-Tan Nguyen [20] proposed a numerical method for simulations
of full three dimensional MHV with moving leaflets in a typical human cardiac cycle. A
cell-centered finite volume method is employed to model incompressible flows in MHV.
The numerical results for laminar and turbulent flows are then validated against
experimental data using Particle Image Velocimetry technique. The schematic diagram of
the experimental setup is shown in Figure 2.1. In the CFD simulations, the flows over the
BMHV at different Reynolds numbers ranging from 350 to 5000 corresponding to the
parabolic laminar flow seen at late systole and during diastole as well as peak systolic
velocity of the cardiac cycle are investigated. The simulations for both Re = 350, Re = 750
and Re = 1000 were validated through comparisons with experimental data. In both
simulations, the most experimental trends and magnitude are captured with reasonable
accuracy. As for the finding, a triple jet structure is observed in the simulations together
with a switching of central orifice jet flow from horizontal axis to vertical axis downstream
of the leaflets and the results are well compared with the experimental data.

Figure 2.1. Schematic diagram of steady flow loop setup

Besides, the paper by S. Annerel [21] discussed on the validation of a recently

developed fluid–structure interaction (FSI) coupling algorithm to simulate numerically the

7
dynamics of an aortic bileaflet mechanical heart valve (BMHV). This validation is done by
comparing the numerical simulation results with in vitro experiments. For the in vitro
experiments, the leaflet kinematics and flow fields are obtained via PIV technique. The
results reveal great similarity in leaflet motion and flow fields between the numerical
simulation and the experimental test. Therefore, it is concluded that the developed
algorithm is able to capture very accurately all the major leaflet kinematics and dynamics
and can be used to study and optimize the design of BMHV.

Figure 2.2. Schematic overview of the experimental setup

Comparison of Previous Studies and Current Study

To summarize, majority of mechanical heart valves used today are bileaflet valves
because they allow the least resistance to flow and the least blood damage. In terms of
numerical solution, FSI is a good approach in hemodynamic analysis of mechanical heart
valve as physical parameters such as wall shear stress and leaflet deformation can be
evaluated. Furthermore, PIV measurement is an excellent method to trace the flow pattern
and measure the velocity of the fluid.

Up until now, there are limited studies that combine both FSI and PVI to study the
hemodynamic of mechanical bileaflet heart valve especially in mitral valve position. For
example, the paper by S. Annerel [21] discussed both FSI and PIV but foucs on the aortic
valve rather than the mitral valve. Next, Vinh-Tan Nguyen paper [20] mentioned about
limitations such as neglecting the hinge mechanism of the valve leaflets and the simulation
was started with the leaflets in fully open position due to meshing issues. Thus, to make a
breakthrough, this research focus on the mitral BMHV, implements a simplified hinge
mechanism and the simulation is started with the leaflets in fully closed position.

8
 : Methodology
The research focuses on the application of mechanical heart valve as an alternative
solution when natural heart valves become dysfunctional. The target audiences of this
research are patients with valvular heart disease, doctors and researchers who are in the
field of cardiology. The research work is meant to be hemodynamic of mechanical heart
valve in the left heart in three dimension using engineering simulation software to analyze
the results. Experiment, too, will be conducted to validate the simulation results.

3.1 Simulation using ANSYS Fluid Structure Interaction (FSI)

Fluid-structure interaction (FSI) is a multiphysics coupling between the laws that
describe fluid dynamics and structural mechanics. This happening is characterized by
interactions between a deformable or moving structure and a surrounding or internal fluid
flow. In the simulation, 2-way FSI is used as the mechanical model and CFD model
analysis results are correlated. A general 2-way FSI setup procedure is outlined in Figure
3.1.

Set up geometry

Set up mechanical model

- Material
-Loads and constraints
-Identify the Fluid-Solid Interface

Set up CFD model

-Flow properties and boundary conditions
- Identiify the Fluid-Solid Interface and specify the
mesh motion model

Set up execution controls and solve

-Time duration and time steps
- Number of coupling iterations per time step
-Convergence criteria

Figure 3.1. General 2-way FSI setup procedure

9
3.1.1 Pre-Analysis
In the Pre-Analysis step, three aspects will be reviewed.

 Governing equations that need to be solved

 Boundary conditions that are applied
 Dynamic mesh that is employed

3.1.1.1 Governing Equations

Before commencing a CFD simulation, it is always good to look at the governing
equations underlying the physics. In this case, although there are additional complexities
such as pulsatile flow and non-newtonian fluids, the governing equations are the same as
any other fluids problem. The most fundamental governing equations are the continuity
equation and the Navier-Stokes equations.

Continuity Equation

𝜕𝜌
+ ∇ ∙ (𝜌v) = 0 Eq. 1
𝜕𝑡

However, since blood is an incompressible fluid, the rate of change of density is zero, thus
the continuity equation above, Eq. 1 can be further simplified in the form below:

∇∙v=0 Eq. 2

Navier-Stokes Equation

𝑑v
𝜌 ( 𝑑𝑡 + v ∙ ∇v) = −∇p + μ∇2 v + f Eq. 3

One thing to note in the Navier-Stokes equation is that the viscosity coefficient, μ is not a
constant but instead a function of shear rate. Blood gets less viscous as the shear rate
increases. This phenomenon is known as shear thinning. Hence, Carreau fluids model is
employed to model the blood viscosity. The mathematical formulation of Carreau model
is as follows:
𝑛−1
𝜇𝑒𝑓𝑓 (𝛾) = 𝜇𝑖𝑛𝑓 + (𝜇0 − 𝜇𝑖𝑛𝑓 )(1 + (𝜆𝛾)2 ) 2 Eq. 4

where

10
𝜇𝑒𝑓𝑓 = effective viscosity

𝜇0 = viscosity at zero shear rate (Pa⋅s)

𝜇𝑖𝑛𝑓 = viscosity at infinite shear rate (Pa⋅s)

material coefficients
𝜆 = relaxation time (s)

n = power index

For the case of blood,

𝜇0 = 0.056 (kg/m⋅s)

𝜇𝑖𝑛𝑓 = 0.0035 (kg/m⋅s)

𝜆 = 3.313 (s)

n = 0.3568

3.1.1.2 Boundary Conditions

Inlet

To the best of our knowledge, mammalian blood flow is pulsatile and cyclic in
nature. Therefore, the velocity at the inlet is not set to be a constant, but instead, in this
case, it is a time varying periodic profile. The pulsatile profile within each period is a
combination of two phases namely the systolic and the diastolic phase. During the systolic
phase, the velocity at the inlet varies in a sinusoidal pattern. The sine wave during the
systolic phase has a peak of 0.5 m/s and a minimum velocity of 0.1 m/s. Assuming a rapid
heartbeat rate of 120 per minute, the duration of each period is 0.5s. This model for pulsatile
blood flow is proposed by Sinnott et, al. [22] A figure of the profile within two periods is
presented in Figure 3.2.

11
 Figure 3.2. Inlet velocity profile in two periods [23]

To illustrate the profile with clarity, a mathematical description is also given below.

𝑣𝑖𝑛𝑙𝑒𝑡 (𝑡) = 0.5 sin[4𝜋(𝑡 + 0.0160236)] 0.5n < n ≤ 0.5n + 0.218

0.1 0.5n + 0.218 < n ≤ 0.5(n+1)

Where n = 0,1,2…

Outlet

The systolic pressure and diastolic pressure of a healthy human is around 120 mmHg and
around 80 mmHg respectively. Thus, taking the average pressure of the two phases, 100
mmHg (around 13332 Pascal) is used as the static gauge pressure at the outlets.

Wall

The easiest boundary condition to determine is the left atrial and ventricular wall. The wall
regions of this model are simply defined and set to “wall”. From a physical viewpoint, the
“wall” condition dictates that the velocity at the wall is zero.

12
3.1.1.3 Dynamic Mesh

Smoothing Method

Diffusion based smoothing is integrated in the dynamic mesh. It tends to produce

better quality meshes than spring-based smoothing and often allows larger boundary
deformations before breaking down. For diffusion-based smoothing, the mesh motion is
governed by the diffusion equation ∇ ∙ (𝛾∇𝑢) = 0 where u is the mesh displacement
velocity. The diffusion coefficient can be used to control how the boundary motion affects
the interior mesh motion. In this simulation, the formulation for the diffusion coefficient,
1
𝛾 is a function of the boundary distance with a formula of 𝛾 = 𝑉 𝛼 where 𝛼 is the diffusion

parameter. In the simulation, diffusion parameter of 𝛼 = 1.4 is applied.

Remeshing Method

Remeshing is also exerted as an insurance because when the boundary displacement

is large compared to the local cell sizes, the cell quality can deteriorate or the cells can
become degenerate. This will invalidate the mesh (for example, result in negative cell
volumes) and consequently, will lead to convergence problems when the solution is
updated to the next time step. To circumvent this problem, remeshing option is activated.
By using remeshing, it agglomerates cells that violate the skewness or size criteria and
locally remeshes the agglomerated cells or faces. If the new cells or faces satisfy the
skewness criterion, the mesh is locally updated with the new cells (with the solution
interpolated from the old cells). Otherwise, the new cells are discarded and the old cells
are retained.

3.1.2 Establishment of CAD Model

Generally, the CAD model consists of two major parts, namely the left heart and
the mechanical bileaflet heart valve. The model is constructed using SolidWorks instead
of using the real left heart which is scanned by 3D scanner. This is because the real scanned
model has lots of uneven surfaces on its irregular geometry which increase the difficulty
to assemble both the left heart and mechanical heart valve together. Moreover, the
complexity of the mesh due to highly irregular geometry will also increase the
computational time. Since the aim of this research is to investigate the effect of different

13
leaflets curvature on blood flow, therefore a simplified model is enough to carry out the
simulation.

The simplified model is constructed using SolidWorks 2016. Loft features are
widely used to create the irregular geometry. The left heart model consists of one inlet and
one outlet and the left atrium and ventricle length are 50mm and 80mm respectively. The
diameter of the outer ring is 31mm. Regarding the leaflets, the leaflets are initially in closed
position where the position is 60˚ from the horizontal axis. The simplified model used in
the simulation is shown in Figure 3.3. Overall, there are five different leaflets curvature
that have been designed. The leaflets curvature modifications are explained in Figure 3.4
and the five different leaflets curvature are shown in Figure 3.5.

Figure 3.3. Simplified CAD model for simulation with predefined dimensions

Figure 3.4. Explanation of leaflets curvature modifications

14
 (a) (b) (c) (d) (e)

Figure 3.5. Different leaflets curvature of the mechanical heart valve. (a) No curved (original), (b)
0.74˚ curved inwards, (c) 0.74˚ curved outwards, (d) 2.22˚ curved inwards, (e) 2.22˚ curved
outwards

The five CAD models are then exported to STEP file format which is suitable to be read
by ANSYS Design Modeler for further processing.

3.1.3 Transient Structural

Before importing the STEP file, the material properties of pyrolytic carbon are
defined as the leaflets of mechanical heart valve are made from pyrolytic carbon. The
general physical properties of pyrolytic carbon are listed below in Figure 3.6.

Figure 3.6. General physical properties of pyrolytic carbon [24]

15
The physical properties that are used in ANSYS simulation are shown in Figure 3.7.

Density = 1850 kg / m3

Young’s Modulus = 24.5 GPa

Poisson’s ratio = 0.22

Elastic limit = 345 MPa

Figure 3.7. Physical properties of pyrolytic carbon that used in ANSYS simulation

3.1.3.1 Geometry
The STEP file is imported to ANSYS Design Modeler in ZX plane. Generally, there
are five parts that make up the whole model namely rod-x, rod+x, leaf+x, leaf-x and heart.
Here, Boolean (subtract) operation is applied to create a cavity as the fluid solid interface
for the fluid domain. The target body will be the heart while the remaining four parts will
be the tool bodies. The geometry imported to the Design Modeler is shown in Figure 3.8.

Figure 3.8. Geometry imported to the ANSYS Design Modeler

16
3.1.3.2 Mesh
Since in transient structural, the focus is the leaflets so the left heart model is
suppressed. The meshing is only done on the leaflets. Here, adaptive size function with fine
relevance center is adopted. In addition, refinement is applied at the 4 edges of the leaflets
as shown in Figure 3.9. The meshed geometry is presented in Figure 3.10.

Figure 3.9. Four edges that are highlighted for refinement

Figure 3.10. The meshed geometry(leaflets)

17
3.1.3.3 Physics Setup

The complex hinge from the exact model is simplified with a pin joint and weak
constraints to control the designated valve leaflet positions. Here, the default contacts are
not suitable to represent the hinge motion. Thus, new joints contacts are established. Here,
the top and bottom faces of the hinge are fixed to the ground. Also, revolute joints are also
defined to allow the leaflets to rotate in z direction. Figure 3.11 and Figure 3.12 shows the
fixed and revolute joints respectively.

Figure 3.11. Face that is selected for the fixed to the ground joint

Figure 3.12. Faces that are chosen for the revolute joints

For the analysis settings, the step end time applied is 0.5s with auto time stepping
disable. The time step is defined by 0.01s interval. About the loads, joints loads are applied

18
to create the rotational motion in z direction. In addition, 18 faces of the leaflets are selected
for the Fluid Solid Interface and shown in Figure 3.13.

Figure 3.13. 18 faces that are selected for Fluid Solid Interface

3.1.4 Fluid Flow (FLUENT)

The geometry that is used previously in transient structural is also used in FLUENT.

3.1.4.1 Mesh

Similarly, in FLUENT the fluid domain which is the left heart is prioritized and the
leaflets are suppressed. Virtual topology is created to merge the surfaces in order to
simplify the geometry. The geometry before and after virtual topology is applied are shown
in Figure 3.14. It is clearly shown that the surfaces are merged after applying virtual
topology thereby optimizing the generated mesh.

Before After

Figure 3.14. Geometry before and after apply virtual topology

19
Next, named selections (inlet, outlet, wall_heart, fluid_domain and fsi) are created. Figure
3.15 displays the faces selected for the respective name.

(a) (b)

(c) (d)

(e)

Figure 3.15. Faces selected for named selection. (a) Inlet, (b) Outlet, (c) Wall_heart, (d)
fluid_domain, (e) fsi

Regarding the mesh, curvature size function with coarse relevance center is applied.
Fine span angle center is also chosen. The mesh quality is justified based on skewness and
orthogonal quality to ensure the mesh of the model is good for the simulation. Figure 3.16

20
shows the range of skewness and orthogonal quality to gauge the quality of the mesh. The
cross-sectional view of the meshing of the model is shown in Figure 3.17.

Figure 3.16. Range of skewness and orthogonal quality to gauge the quality of mesh

Figure 3.17. Cross-sectional view of the meshing of the model

3.1.4.2 Physics Setup

In FLUENT, the time is changed to transient since we wish to observe the blood
flow pattern over time. Laminar viscous model is also chosen. For the material, blood with
a density of 1060 kg/m3 is applied. The viscosity is described using the carreau fluid model
with predefined parameters as mentioned in section 3.1.1.1 Governing Equations.

21
 For the boundary conditions, the inlet is set to velocity inlet. As mentioned in
section 3.1.1.2 Boundary Conditions, blood flow is pulsatile so a udf as shown in Appendix
1 is used to describe the periodic profile at the inlet. At the same time, the outlet is set to
pressure-outlet with a gauge pressure of 13332 Pa. The remaining named selection (fsi,
wall_heart and fluid_domain) are set to wall.

Besides, dynamic mesh is implemented. Smoothing and remeshing mesh methods

are activated. The parameters of both mesh methods are as shown in Figure 3.18. Next, to
define the dynamic mesh zones it is important to defined the fsi as the system coupling and
the remaining as stationary.

Figure 3.18. Parameters for smoothing(left) and remeshing(right) mesh method

Figure 3.19 shows the mesh distribution of the left heart model when the leaflets
are fully closed and opened. It is observed that the deformation of the mesh from closed to
opened position is not large. Thus, it does not encounter negative cell volume detected
22
issue which means that the element does not have a big distortion. Not big enough to do
some vertex, or edges, to penetrate an opposite face resulting in a negative volume.
Therefore, the dynamic mesh used is appropriate.

(a) (b)
Figure 3.19. Mesh distribution of the left heart model. (a) Fully closed, (b) Fully opened

For the reference values, the area is changed to follow the body area in design
modeler which is 0.023565 m2. This value is obtained by using the analysis tool to study
the entity information which will then display the details of the simulation model. The
reference values of density and velocity are changed to 1060 kg/m3 and 0.1 m/s respectively.

Upcoming will be the pressure-velocity coupling method. Here, the coupled

scheme is applied. The pressure-based coupled algorithm obtains a more robust and
efficient results. Also, the hybrid initialization is employed as solution initialization. Lastly,
the time step size is set to 0.01s with 50 number of time steps and max iterations per time
step is set to 30. However, it is important to but note that the system coupling’s time step
size will override FLUENT predefined time step value.

3.1.5 System Coupling

System coupling is the most vital part in fluid structural interaction simulation as it
links both the transient structural and FLUENT together to provide a final solution. Physics
are coupled by passing loads across the fluid structure interfaces where CFD transfer forces
on structure surfaces to FEA while FEA transfer displacements of solid structures to CFD.
Individual physics are solved separately and then coupled sequentially or simultaneously
until equilibrium is achieved. Figure 3.20 shows the flow chart on how the system coupling
works in ANSYS.

23
 Figure 3.20. Flow chart showing how the system coupling works in ANSYS

In ANSYS, the two way FSI is set up as shown in Figure 3.21. The setup of transient
structural and FLUENT are linked together to the system coupling for the solution purpose.

Figure 3.21. Two way FSI setup in ANSYS

In the system coupling, the end time is set to 0.5s with step size of 0.01s. Minimum
iteration is set to 1 whereas maximum iterations are set to 30. Next, data transfers are
established between Fluid Solid Interface in Transient Structural and fsi in FLUENT. Here,
the surface of the fluid system around the leaflets transfers force to the surface of the

24