Applied Biomedical Engineering

AMME4981

Lecture 5
Finite Element Modelling

Course Web
http://www.aeromech.usyd.edu.au/people/academic/qingli/AMME4981.htm
Contents

 Multiphysical modelling in biomedical engineering

 Prosthetic failure (Fracture and Fatigue)

 Ansys Workbench
FE Multiphysical Modelling

 Static analysis – stress, strain, displacement

 Dynamic analysis – Frequency, mode shape

 Acoustic analysis – Sound pressure

 Thermal analysis – Temperature, heat flux

 Fluid analysis – Velocity, shear stress

 Electrical analysis – Electrical field intensity

 Magnetic analysis – magnetic flux, magnetic potential

 Coupled analyses – Thermal-Structural, Fluid-Structural,

Electrical-magnetic, etc
I. Dynamic Analysis – Osseointegration Study

Natural Frequency Analysis of Osseointegration for Trans-Femoral Implant

Shao et al, Annals of Biomedical Engineering 35:817–824, 2007

Mechanism

 Hypothesis:
Different osseointegration degree leads to different interfacial condition, thereby
different natural frequency. Different silicone rubber interfaces with the curing
time are used to simulate (mimic) osseointegration process.

 Method:
 Use the resonant characteristics of the implant system to determine the
changes in stability as a reflection of boundary condition of the implant. With a
small mechanical excitation,
 Vibration responses of the trans-femoral implant to a small mechanical
excitation are measured using an accelerometer
 The vibration signal will be analyzed using Fast Fourier Transform (FFT)
software to obtain the fundamental natural frequency of the implant system.

 Result:
In-vitro study was conducted using different silicone rubbers to simulate the
interface condition. The result showed that a high NF corresponded to a high
elastic modulus of the interface material between the implant and bone.
Apparatus in-vitro and in-vivo

In-Vitro Apparatus In-vivo Apparatus

Curing time of silicon rubber Real osseointegration measurement

Shao et al, Annals of Biomedical Engineering 35:817–824, 2007

Result – Natural Freq vs Remodelling Time

The longer the curing time, the

higher the natural frequency,
thereby the better the
osseointegration and more
stable prosthetic outcome

The longer the healing time, the

higher the natural frequency,
thereby the better the
osseointegration and more
stable prosthetic outcome
Result –Freq vs Interfacial Young’s Modulus

The longer the curing time, the

higher the interfacial Young’s
modulus, the higher the natural
frequency, thereby the better
the osseointegration and more
stable prosthetic outcome

Modelling Scheme
Blood layer with a Young’s
modulus between 0.1 and
0.7MPa
Implant
Host
Bone
Osseointegration in Dental Implants

Peri-Implant Layer
E1=0.05 – 0.6GPa

E2=16GPa

E3=0.6GPa

A peri-implant interfacial area, up to 1 mm external to the implant thread, was

carefully modelled with the aim of differentiating the anatomical region that was
damaged by implant placement and whose mechanical properties changed during
the integration process. In fact, tissue damage phenomena because of the
mechanical drilling process are unavoidable and there is strong evidence of bone
tissue activity in the healing phase after the implant placement, resulting in new bone
deposition and structural rearrangement.
Dynamic Result for Dental Implant

Relation between 1st

natural frequency vs
interfacial Young’s
Osseointegration modulus – better
osseointegration

Cortical bone remodeling

Nodal displacement 16000000000

Young's modulus of cortical bone (GPa)

point 1
15500000000

Nodal displacement 15000000000

Point 2

14500000000

14000000000

13500000000

13000000000
2

10

12

14

16

20

24

28

32

36

44

48
k

k
ee

ee

ee

k
ee

ee

ee

ee

ee

ee

ee

ee

ee

ee

ee
w

w
Time (week)
Boen remodelling
Young’s Modulus vs healing Time
II. Fluid Flow – Pulse Blood Flowing in Artery

CFD-Structural
Ansys – From Leap Australia
Coupled Problem
Biomedical Example – Particle Flow in Lung

Ansys – From Leap Australia

Biomedical Example – Air Flow in Pulse

Flow-Structural coupling problem

Ansys – From Leap Australia

Biomedical Example – Valve Sheldon
 III. Electromagnetic modelling

FEA model Specific Absorption Rate (SAR) Temperature Change

Electromagnetic radiation effect of mobile phone on possible consequences of on health.

Modelling – Adaptive Meshing

Dense Mesh

Modeller – Ansys Workbench

IV. Dynamic-Flow-Electrical Cochlear Mechanics

Middle Ear

External Ear Inner

Ear
 External Ear (E) comprises the pinna and acoustic meatus. It funnels sound
vibrations to the middle ear.
 Middle Ear (M) comprises three tiny bones (ossicles): malleus, incus and
stapes. Its function is to transform air-born vibrations impinging on the
tympanic membrane into oscillations of the fluid filling the inner ear.
 Inner Ear (I) - Cochlear
The Middle Ear as a Mechanical Transformer
The middle ear is a full-fledged mechanical
energy transformer. The three ossicles (malleus,
incus and stapes) work like a lever system that
increases the force transmitted from the
tympanic membrane to the stapes by decreasing
the ratio of their oscillation amplitudes. The
footplate of stapes acts like a small piston on the
cochlear fluid through a membraneous
connection that seals the oval window of the
cochlea. The buckling motion of the tympanic
membrane decreases the velocity two-fold an
increases the force two-fold, changing the
impedance ratio four-fold. Thanks to the large
surface ratio between tympanic membrane and
oval window (about 35) and the ossicle system
lever gain (about 1.32), the forward impedance
gain is about 30 dB. Filtering effects due to
resonances of the middle ear cavity and
mechanical parameters of the ossicle system
produce a peak between 1 and 2 kHz.
Basilar Membrane as a Thin Elastic Fibers
The basilar membrane is internally formed by thin elastic fibers tensed across
the cochlear duct. The fibers are short and closely packed in the basal region,
i.e. close to the stapes, and become longer and sparse proceeding towards the
apex of the cochlea, where the basilar membrane ends in a foramen that joins
the two partions of the spiral canal (see Anatomy). Being under tension, the
fibers can vibrate like the strings of a musical instrument. Traveling waves peak
at frequency-dependent locations, higher frequencies peaking closer to more
basal location. Peak position is an exponential function of input frequency
because of the exponentially graded stiffness of the basilar membrane. Part of
the stiffness change is due to the increasing width of the membrane and part to
its decreasing thickness.
 Cochlear Hydrodynamics – 2D Modelling

Fluid

A longitudinal section of the uncoiled cochlea is represented with vertical dimension

expanded by about 3 times. A traveling wave elicited by a 3 kHz tone is shown as a solid
red line displacing the basilar membrane (unbroken black line) from its resting position (the
wave amplitude is magnified about 106 times for clarity) Arrows around wave peaks indicate
the direction of local fluid flow. The fluid mass affects the dynamics of the basilar
membrane, loading its different parts by amounts that depend upon the local wave length.
Notice the progressive shortening of the wave length up to a critical point beyond which
both the basilar membrane and the fluid keep at rest. The inset shows a cross section of the
spiral canal, showing that the basilar membrane is laterally clamped across the duct and
supports the organ of Corti, that hosts two types of sensory hair cells: inner hair cells, that
transmit signals to the acoustic nerve, and outer hair cells that provide mechanical
amplification to the basilar membrane motion
In Cochlear – Hari Cells

Static Sounded

The animation at left visualizes the activation of an IHC by

the fluid viscous drag applied to its stereocilia by the
oscillation of the tectorial membrane. Outer hair cells are
the target of abundant efferent innervation and possess a
unique type of motility. They convert receptor potentials
into cell length changes at acoustic frequencies. The
animation at right visualizes the activation of the outer hair
cell motor driven by the motion of the tectorial membrane
into which the tips of the tallest stereocilia are inserted. A
second class of sensory receptors, the outer hair cells
couple visco-elastically the reticular lamina to the basilar
membrane through their supporting Deiters' cells.
V. Fracture Modelling

Mode I
(Tension, opening)

Mode II
(In-Plane Shear, Sliding)

Mode III
(Out-Of-Plane Shear,
Tearing)
Fracture Mechanics – Griffith Theory
 Griffith's Crack Theory: strain energy release rate

 a


For the simple case of a thin rectangular plate with a crack perpendicular
to the load Griffith’s theory becomes:

where G is the strain energy release rate, σ is the applied stress, a is

half the crack length, and E is the Young’s modulus. The strain energy
release rate can otherwise be understood as: the rate at which energy is
absorbed by growth of the crack.

However, we also have that:

If G ≥ Gc, this is the criterion for which the crack will begin to propagate. f
is the critical stress for failure of material
 Fracture Mechanics – Irwin's Theory

 Irwin's modified Griffith crack theory: fracture toughness

Eventually a modification of Griffith’s theory emerged from this work; a term called
stress intensity replaced strain energy release rate and a term called fracture
toughness replaced surface energy. Both of these terms are simply related to the
energy terms that Griffith used:
KI    a
EGc
Plane stress: K c  EGc Plane strain: K c 
1  2
where KI is the stress intensity, Kc the fracture toughness, and ν is Poisson’s ratio.
 It is important to recognise the fact that fracture parameter Kc has different values
when measured under plane stress and plane strain
 Fracture occurs when KI ≥ Kc. For the special case of plane strain deformation, Kc
becomes KIc and is considered a material property.
 The subscript I arises because of the different ways of loading a material to
enable a crack to propagate. It refers to loading via Mode I - the most common form
of loading:
Stress Intensity – Geometric Effect

 Dimensionless
We must note that the expression for stress intensity KI will be different
for geometries other than the center cracked plate. Consequently, it is
necessary to introduce a dimensionless correction factor, Y, in order to
characterise the geometry. We thus have:
K I  Y  a
where Y is a function of the crack length and width of sheet given by:

a  a 
Y    sec 
W  W 

 a
 W
VI. Fatigue Analysis

Material Fatigue Experiment Ensure the hip prostheses against

static, dynamic and fatigue failure.

How to conduct fatigue analysis

Step 1. Conduct static analyses
under body load.
Step 2. Conduct dynamic
analyses were performed under
walking load.
Step 3. Fatigue behavior of stem
shapes was predicted using
ANSYS Workbench for given
material fatigue data (S-N curve).

Senalp et al, Materials and Design 28 (2007) 1577–1583

VI. Fatigue Modelling

Mean stress

Alternating
stress

Goodman

Solderberg
2
Gerber

Su – Endurance limit
Sy – Yield Strength

Senalp et al, Materials and Design 28 (2007) 1577–1583

VI. Fatigue Result
Under Static Load Under Dynamic Load

Senalp et al, Materials and Design 28 (2007) 1577–1583

 VII. Ansys Workbench – A Design Tool

Download tutorial material from

http://www.aeromech.usyd.edu.au/people/academic/qingli/AMME4981.htm

 Design Modeler Evaluation Guide – A Quick Tutorial

 static Analysis

 Modal Analysis
Summary

 FE Multiphysics Modelling

 Prosthetic Failure (Fracture and Fatigue)

 Ansys Workbench

Task in Week 5:
 Read two papers of
Shao et al, Annals of Biomedical Engineering, 35:817–824, 2007
Senalp et al, Materials and Design 28:1577–1583, 2007
 Attempt Ansys modelling in classic GUI and/or Workbench
 Group discussion of project on modelling