Proceeding Of International Conference On Recent Innovations In Science, Engineering And Technology ICRISET14,

2nd & 3rd April 2014.

MODELLING AND ANALYSIS OF PART SEGMENT OF HUMAN VERTEBRAL SPINE

F.R.Femin1
Assistant Professor, Department of mechanical engineering
Immanuel Arasar JJ College of Engineering, Nattalam, Marthandam K.K District
frfemin1983@rediffmail.com
ABSTRACT
Human Spine is an important part of a human
being. Spine deformity refers to the group of conditions in
which the spine exhibits abnormal three-dimensional
curvature. The spine is a very complex structure, and many
of the models are simplified and idealized because of the
complexity and uncertainty in the geometry, material
properties and boundary conditions of these problems. This
type of modeling simplification is not peculiar to spinal
modeling problems. Indeed, the idealization is often strength
when there is such uncertainty and variation between one
individual and another, allowing cause-effect relationships to
be isolated and fully explored, and the inherent variability of
experimental tests to be eliminated. The main objective of
this current work is

To understand the process and methodology of

conversion of CT scan data of a human spine to a
CAD model and from there to be used in a FE
Analysis

To validate the FE model created with the

experimental data available in the literature
publications

To understand the behaviour of the spine model

under defined properties, its meshing behaviour and
going on to study the displacement and stress profile
under biomechanical loading conditions with and
without degeneration.

on development and the evolution of our skeleton and the

constituents of our body.
Many biomechanics work in a competitive research
environment which does not always permit them to investigate
problems of their own choosing. Research funding is replacing
discovery as the major determinant of academic career
progression, and funding is increasingly directed towards
fewer and larger research groups. Collaboration and consensus
are now valued more highly than diversity and debate, so it is
important for young scientists to investigate the right
research questions in the right company. Against this
background, the authors offer a brief review of current trends
in spine research, and suggest which of them hold the most
opportunity for biomechanics. In such a task, it is difficult to
avoid personal bias, to see every problem as a nail when your
only tool is a hammer. However, with research interests
ranging from mechanical tissue testing to patient rehabilitation,
the authors hope to avoid undue subjectivity. There is not space
in this short review to consider specific spinal disorders such
as idiopathic scoliosis or spondylolysthesis. Instead, the review
concentrates on the central problem of age-related spinal
degeneration (spondylosis) and associated pain. Some
recent publications have suggested that biomechanics is largely
irrelevant to this condition; because genetic inheritance largely
determines Intervertebral disc degeneration as well as much of
the resulting pain, and psychosocial characteristics dominate
all aspects of back pain behaviour. However, the evidence
summarized below suggests that this is an over-reaction, and
those excessive mechanical loading remains one of the most
important and preventable causes of spinal degeneration and
pain.
COMPONENTS OF HUMAN SPINE

Various basic levels of disc degeneration under

biomechanical loading needs to be analysed to find
out the implications on the behaviour of the spine
especially the displacement of facet joint.

INTRODUCTION
Biomechanics is the theory of how tissue, cells,
muscles, bones, organs and the motion of them and how their
form and function are regulated by basic mechanical
properties. The purpose of this study is to illustrate the
potential that a combination of knowledge from biology and
engineering can have on the understanding of the development,
maintenance and repair of the skeleton and on the clinical
management of bone diseases. Darwin stated in his treatise, On
the Origin of Species, that the mechanical loading of the
environment and the circumstances of life have a huge impact

The purpose of the spine is to help us stand and sit

straight, move, and provide protection to the spinal cord. The
human spine has 7 Cervical vertebras (C), 12 Thoracic vertebra

95
Department of EEE & Mechanical Engg., Immanuel Arasar JJ college of Engineering

Proceeding Of International Conference On Recent Innovations In Science, Engineering And Technology ICRISET14,
2nd & 3rd April 2014.

(T), 5 Lumbar vertebra (L), and 5 Sacral vertebra (S).

Generally, these are numbered in the following manner: the
first cervical vertebra is C1 and the last cervical vertebra is C7.
The first thoracic vertebra is T1 and the last thoracic vertebra is
T12. Similarly, the first lumbar vertebra is L1 and the last
lumbar vertebra is L5. After the L5 vertebra is the sacrum and
although there are 5 sacral vertebra, there typically are no discs
intervening here. Between the second cervical vertebra and the
first sacral vertebra are intervening discs. The discs are
structures made out of cartilage and they provide cushion
between each vertebral segment. The discs are then referred to
by the vertebra which borders them: the disc between the fifth
lumbar vertebra and the first sacral vertebra is the L5, S1 disc,
the disc between the fifth and sixth cervical vertebra is the C5,
6 disc. Please note, that there are individuals born with either
one extra or one less vertebra and this may alter the numbering
system at times.
Further, each vertebra joins the above and below
vertebra with many different types of joints. Typically each
vertebra has 4 facet joints: a right and left superior and inferior
facet. The cervical vertebra also has 2 joints which are called
uncovertebral joints. Between the first and second cervical
vertebra, there is a special joint which allows rotation of the
head. The first and second vertebra are thus extremely unique
and do not follow most of the typical anatomical rules.
Since the spine helps us to stand straight, its shape is
designed to carry the weight of the body and distribute it
straight down through the pelvis. Each individual segment of
the spine also has its unique curvature. The cervical curvature
is called the cervical lordosis (backward bend), the thoracic
curvature is called the thoracic kyphosis (forward bend), and
the lumbar curvature is called the lumbar lordosis (backward
bend).

Fig 1.2 Lumbar Spine Front View

Fig 1.3 Lumbar Spine Side View

SPINE BIOMECHANICS:
Current trends in spine research are reviewed in order
to suggest future opportunities for biomechanics. Recent
studies show that psychosocial factors influence back pain
behaviour but are not important causes of pain itself. Severe
back pain most often arises from intervertebral discs,
apophyseal joints and sacroiliac joints, and physical disruption
of these structures is strongly but variably linked to pain.
Typical forms of structural disruption can be reproduced by
severe mechanical loading in-vitro, with genetic and age
related weakening sometimes leading to injury under moderate
loading. Biomechanics can be used to quantify spinal loading
and movements, to analyse load distributions and injury
mechanisms, and to develop therapeutic interventions. The
authors suggest that techniques for quantifying spinal loading
should be capable of measurement in the field so that they
can be used in epidemiological surveys and ergonomic
interventions. Great accuracy is not required for this task,
because injury risk depends on tissue weakness as much as
peak loading. Biomechanical tissue testing and finite-element
modelling should complement each other, with experiments
establishing proof of concept, and models supplying detail and
optimising designs. Suggested priority areas for future research
include: understanding interactions between intervertebral
discs and adjacent vertebrae; developing prosthetic and tissueengineered discs; and quantifying spinal function during
rehabilitation. Mechanobiology has perhaps the greatest
future potential, because spinal degeneration and healing are
both mediated by the activity of cells which are acutely
sensitive to their local mechanical environment. Precise
characterisation and manipulation of this environment will be a
major challenge for spine biomechanics.
Current Trends in Spine Research
1. Back pain is associated with spinal degeneration
2. mechanical loading can precipitate spinal injury
3. Spinal degeneration can represent a cell-mediated
response to injury
4. Functional pathology
PROBLEM FORMULATION

96
Department of EEE & Mechanical Engg., Immanuel Arasar JJ college of Engineering

Proceeding Of International Conference On Recent Innovations In Science, Engineering And Technology ICRISET14,
2nd & 3rd April 2014.

As discussed in the introduction chapter and in the

literature review there are well brought out constraints in the
modeling of spine due to its physical makeup having diverse
materials with different properties as shown in Fig (3.1)

Fig 3.1 Various components of spine

3.1. MATERIAL PROPERTIES AND ELEMENT TYPES
FOR MESHING IN SPINE
In general, the material types and properties were
chosen as a compromise of accuracy and time-efficient
analysis.
3.1.1. Vertebral Body
All trabecular bone and posterior parts were modeled
by four-node three-dimensional tetrahedral solid elements
(TETRA4). The cortical bone and the cortical endplates
consisted of a shell layer 1- and 0.5-mm-thick surrounding the
trabecular core, respectively. The cortical shell was modeled by
three-node triangular thin shell elements (SHELL3). Both
element types were chosen, since automated mesh generation
was available in the FEA program, thus minimizing the timeconsuming manual meshing. By keeping a uniform mesh
density for mesh generation of both solid elements and shell
elements a total bonding situation was obtained, thus
increasing accuracy. The material properties for cortical bone,
trabecular bone and bony posterior parts were considered
isotropic and linear, thus minimizing the analysis time.
3.1.2. Ligaments
The ligaments were applied to the model manually.
The bony insertion points were connected by two node3Duniaxial elements (TRUSS3D).The directions of the elements
were orientated along the fiber direction of the ligaments.
Since the tension forces were relatively low while at rest, the
ligaments were assumed to be initially unstressed, and were
defined to be active in tension only the nonlinearity of the
ligaments maintained stability to the FEA model, while giving
adequate displacement responses.

Fig 3.2 Ligaments

3.1.3. Facet Joints
The two articular processes of the adjacent vertebrae,
representing the two halves of the lumbar facet joint, were
adjusted by extrusion of the surfaces representing the facet,
and thereby making the articular surfaces congruent, placing
them with a gap of 1 mm. The inclination angle of the facets
was 64.40 to the anatomical horizontal plane and within the
limits of inclination angle of the lumbar vertebral region. The
effect of the compliant facet cartilage layer was incorporated
by TRUSS3D elements active in compression only with an
initially low modulus of elasticity in compression reaching a
maximum value in range of that of the posterior elements as
the gap disappeared and zero modulus of elasticity in tension.
For each facet joint the elements used were 3-D nonlinear
node-to-node contact elements (GAP). No friction was
assumed in the facet joint.
3.1.4. Intervertebral Disc
The disc consisted of the nucleus pulposus and the
annulus fibrosis. The nucleus was considered incompressible
[8] and was modeled as ten-node 3-D tetrahedral solid
elements (TETRA10) to increase accuracy in the disc, due to
the large displacement responses in the region during analyses.
The annulus consisted of four layers of fibers with a volume
percentage of 16% and with a mean angle of 29. The fibers
were modeled as two-node uniaxial elements (TRUSS3D), and
modified to incorporate nonlinearity.
The fibers were embedded in a ground substance with
three layers of ten-node three-dimensional tetrahedral solid
elements (TETRA10). The overall cross-sectional area
occupied by the nucleus was within the range of 30%60% of
the total disc volume. The disc was created geometrically with
an initial outward disc bulge of 0.6 mm taken for the disc
midplane in comparison with the adjacent endplates. Material
properties for the vertebral bodies, posterior parts and
intervertebral discs were linear and isotropic. Material
properties for the ligaments, annulus fibers, facet articulation,
and cartilage were nonlinear
3.1.5. Spine material properties
Component

Youngs modulus, E

Poissons

97
Department of EEE & Mechanical Engg., Immanuel Arasar JJ college of Engineering

Proceeding Of International Conference On Recent Innovations In Science, Engineering And Technology ICRISET14,
2nd & 3rd April 2014.

name
Cortical bone
Cancellous bone
Posterior bone
Endplate
Disc Nucleus
Disc Annulus
Annulus fiber

(MPa)

ratio

12000
100
3500
12000
6.0
4.2
500
Table 3.1

0.3
0.2
0.25
0.3
0.499
0.45
0.3

3.1.6. Geometric Generation

The model generated was a lumbar disc-body unit.
The geometrical surface model of each vertebra consisted of a
triangulated surface mesh. This surface mesh was generated
from computed tomography (CT) images from the Visible
Human Project. The CT slice images had a slice distance and
thickness of 1 mm and a pixel size of 0.33 mm. The procedure
consisted of a parallel correction of the longitudinal
geometrical axis of the CT scan to the longitudinal axis of the
spine. Two-dimensional (2-D) contours were then drawn
manually on an enlarged version of the CT slices with the
visualization software Mvox, and fitted to match the general
anatomical dimensions of the region within one standard
deviation (SD) of normal materials of previous studies. Finally,
the 3-D reconstruction was created from the 2-D parallel
contours by the Nuages Package using Delaunay triangulation.
The 3-D surface mesh was converted into DXF format from
Mvox and was readable by the COSMOSM FEA software.
The sensitivity in FEA results as stresses and strains
were examined by comparing differences in contour points
distances (CPD) for a voxel similarity-based and ten geometric
marker-based segmentation models. The marker-based models
with maximum and minimum average differences in CPD were
converted into FEA models, and the computed finite-element
responses showed differences within acceptable limits.
The sensitivity in FEA results as stresses and strains
were examined by comparing differences in contour points
distances (CPD) for a voxel similarity-based and ten geometric
marker-based segmentation models. The marker-based models
with maximum and minimum average differences in CPD were
converted into FEA models, and the computed finite-element
responses showed differences within acceptable limits.
3.2. IN VITRO MODEL
Testing on cadaveric specimens was done to validate
the FEM because there is no data available in the literature
using human subjects. Ten human cadaveric lumbar spines
from L1 to Sacrum (seven male, three female, mean age 58
years, range 4173 years) were compressed to 1200N in the
neutral posture using the follower load technique. Before
testing, plane radiographs (GE OEC 9800 Digital Flouroscopy
Machine, GE Medical Systems Inc., Waukeshe, WI) were
taken to rule out bony abnormalities and MRI (Hitachi Airis II,
Hitachi Medical Systems, Twinsburg, OH) was performed to

rule out significant soft-tissue abnormalities and to ensure that

each specimen had healthy discs of grade II or better as
determined by an orthopedic surgeon. No preconditioning of
the in vitro specimens was done, other than that done while
optimizing the follower load path as described. During testing,
lateral radiographs were taken before compression (0 N), and
with a 1200N follower load. The heights of each disc under
both loads were measured using ScionImage (Scion Corp.,
Frederick, MY) and taken as the average of the anterior,
posterior, and middle disc heights. In the experimental set up it
was possible to apply compressive pre-load only along with
flexionextension moments. The ROM at each of the five
levels was measured for moments of 8Nm in flexion and 6Nm
in extension without pre-load and with an 800N follower preload. A larger moment was applied in flexion than extension
because the ROM of the spine is larger in flexion than
extension. An 800N pre-load was applied to represent the
average load placed upon the lumbar spine while standing and
to limit the potential for specimen damage during testing.
Lateral bending moments of 76Nm and torsional moments of
74Nm were applied to the cadaver lumbar spines without
preload. As a consequence of the methods used to apply the
moment loads in vitro, a small axial load of approximately 14N
was added to the spine, which is negligible compared to the
pre-load used for the analyses. The flexionextension and
lateral bending rotation of each vertebral body due to the
moment load was measured using bi-axial inclinometers
(Model 902-45, Applied Geomechanics, Santa Cruz, CA). The
torsion of each vertebral body due to the moment load was
measured using an optoelectronic motion measurement system
(Optotrak, NorthernDigital, Waterloo, Ontario). All angular
measurements were made to within 0.11. The total segmental
motion of each motion segment, taken as the sum of the two
corresponding motions in each plane (i.e. flexionextension,
right+left lateral bend, and right+left torsion) was calculated.

Fig.3.3. Lateral (A) and anteriorposterior (B) views of the

in vitrofollower load technique.
The main objective of this current work is

98
Department of EEE & Mechanical Engg., Immanuel Arasar JJ college of Engineering

Proceeding Of International Conference On Recent Innovations In Science, Engineering And Technology ICRISET14,
2nd & 3rd April 2014.

To understand the process and methodology of

conversion of CT scan data of a human spine to a
CAD model and from there to be used in a FE
Analysis

To validate the FE model created with the

experimental data available in the literature
publications

To understand the behaviour of the spine model

under defined properties, its meshing behaviour and
going on to study the displacement and stress profile
under biomechanical loading conditions with and
without degeneration.

Various basic levels of disc degeneration under

biomechanical loading needs to be analysed to find
out the implications on the behaviour of the spine
especially the displacement of facet joint.

.RESULTS AND DISCUSSIONS

The FEM model that was generated need to be
validated. The result from the existing published literature
needs to be compared with the result that was achieved in this
work. The displacement of the L4-L5 vertebra under 400N,
800N&1200N with the boundary conditions as mentioned in
the literature Novel model to analyze the effect of a large
compressive follower pre-load on range of motions in a lumbar
spine Ragu et al was compared (Fig 5.47). A threedimensional non-linear lumbar spine FEM has been developed
that can be used to evaluate the full kinematics response of the
lumbar spine under various levels of degeneration.

5.1. Conversion of CT Scan Data To Cad Model

Fig 5.1.Conversion of CT Scan data to CAD model

5.2. Finite element analysis of l4-l5 vertebra using in Ansys
5.2.1 Scan FE model exported to ANSYS

Fig 5.2. Scan FE model exported to ANSYS

This model of spine was created in ScanFE. ScanFE
provides more suitable platform for creating complicated
geometries. ScanFE helped in creating different material
boundaries for this bone structure. The multiple colours have
shown in the above figure shows different material properties.
The process of creating different material geometries is
difficult in ANSYS. There are four materials in this model (fig
5.2) Cortical born, Cancellous bone, Nucleus pulposus and
Annulus Fibrosus.

5.2.2 Displacement of L4-L5 Vertebra with 1200N,

800N&400N

99
Department of EEE & Mechanical Engg., Immanuel Arasar JJ college of Engineering

Proceeding Of International Conference On Recent Innovations In Science, Engineering And Technology ICRISET14,
2nd & 3rd April 2014.

Fig 5.3. Boundary condition with compressive load

The ANSYS figure shown above represents
applications of boundary conditions as compressive load over
the vertebra. The loads are 400, 800&1200N. The bottom side
of the vertebra was fully contained with zero degree of
freedom.

Fig 5.6 Displacement in 1200N

The above figure shows a displacement of
1.198mm.The extension of outward projection slightly more
than 800N loading condition
DISCUSSIONS
A three-dimensional non-linear lumbar spine FEM
has been developed that can be used to evaluate the full
kinematics response of the lumbar spine under various levels
of degeneration. As seen in the Fig 5.48 the displacement is
found to be1.19mm which agrees with the above mentioned
literature.

Fig 5.4. Displacement in 400N

The above figure shows a displacement of 0.306 mm
when load of 400N is applied on the top side of the vertebra.

Fig 5.47 Load Vs displacement

The above graphs shoes a various in displacement
conditions while the compressive loads of 400,800&1200N are
applied on it. The 1200N shows maximum displacement of
1.2mm
Fig 5.5 Displacement in 800N
The above figure shows a displacement of 0.90mm.
The disc selection compressed and an outward projection has
happened.

100
Department of EEE & Mechanical Engg., Immanuel Arasar JJ college of Engineering

Proceeding Of International Conference On Recent Innovations In Science, Engineering And Technology ICRISET14,
2nd & 3rd April 2014.

5.

Ian A.F. Stokes, PhD and Mack Gardner-Morse,

MS.Muscle Activation Strategies and Symmetry of
Spinal Loading in the Lumbar Spine With
Scoliosis.2004

6.

Sairyo K, Goel VK, Masuda A, Vishnubhotla S,

Faizan A, Biyani A, Ebraheim N, Yonekura D,
Murakami R, Terai T., Three-dimensional finite
element analysis of the pediatric lumbar spine.
Part I: Pathomechanism of apophyseal bony ring
fracture.Eur Spine J. 2006 Jun; 15(6):923-9. Epub
2006 Apr 14

7.

Sairyo K, Goel VK, Masuda A, Vishnubhotla S,

Faizan A, Biyani A, Ebraheim N, Yonekura D,
Murakami R, Terai T. , Three dimensional finite
element analysis of the pediatric lumbar spine.
Part II: biomechanical change as the initiating
factor for pediatric isthmic spondylolisthesis at the
growth plate.

8.

Sairyo K, Goel VK, Vadapalli S, Vishnubhotla SL,

Biyani A, Ebraheim N, Terai T, Sakai T.
Biomechanical comparison of lumbar spine with or
without spina bifida occulta. A finite element
analysis. Spinal Cord. 2005 Nov 29; 2004

9.

Zheng-Cheng Zhong, Shun-Hwa Wei, Jung-Pin Wang,

Chi-KuangFeng, Chen-Sheng Chen and Chung-huang
Yu, Finite element analysis of the lumbar spine
with a new cage using a topology optimization
method ,Medical Engineering & Physics, Volume 28,
Issue 1, January 2006, Pages 90-98 ,

Fig 5.48 Comparison of Literature

In vitro, Dr. Raghu result and our result disc

compression with 1200N follower load. Compression predicted
by our result compares well with in other two results.
This validates the methodology adopted for the
conversion of CT scan data to FEM solid model procedure. As
the methodology has been validated the same procedure can be
adopted to study the degenerated disc as well. In the table
maximum deformation, maximum vonmisess stress and
maximum vonmisess strain are compared in three levels for
different disc deformation condition when disc degeneration
are in level 1, level 2 and level 3.the above explanation shows
that the movement of the human body is restricted when disc
degeneration is high.

REFERENCE:
1.

Goto K, Tajima N, Chosa E, Totoribe K, Kuroki H,

Arizumi Y, Arai T. Mechanical analysis of the
lumbar vertebrae in a three-dimensional finite
element method model in which intradiscal
pressure in the nucleus pulposus was used to
establish the model. J Orthop Sci. 2002;7(2):243-6

2.

Christian Wong*, P. Martin Gehrchen, TronDarvann,

and Thomas Kir, Nonlinear Finite-Element
Analysis and Biomechanical Evaluation of the
Lumbar Spine. IEEE Transactions On Medical
Imaging, vol. 22, no. 6, june 2003.

3.

R.N. Natarajan *, J.R. Williams, G.B.J. Andersson.

Finite element model of a lumbar spinal motion
segment to predict circadian variation in stature.

4.

Mack G. Gardner-Morse*, Ian A.F. Stokes.

Structural behavior of human lumbar spinal
motion segments.2004 Journal of Biomechanics

10. Yao J, Turteltaub SR, Ducheyne P., A threedimensional nonlinear finite element analysis of
the mechanical behavior of tissue engineered
intervertebral discs under complex loads.
Biomaterials. 2006 Jan; 27(3):377-87. Epub 2005 Sep
15.
11. Shirazi-Adl, Analysis of large compression loads on
lumbar spine in flexion and in torsion using a
novel wrapping element, Journal of Biomechanics,
Volume 39, Issue 2, 2006, Pages 267-275 ,A.
12. Raymond Y. W. Lee, PhD; Bonnie Y. S. Tsung, MPhil;
Pin Tong, PhD; John Evans, PhD, Bending stiffness
of the lumbar spine subjected to posteroanterior
manipulative force,Journal of Rehabilitation
Research & DevelopmentVolume 42, Number 2,
Pages 167174, March/April 2005.

101
Department of EEE & Mechanical Engg., Immanuel Arasar JJ college of Engineering

Proceeding Of International Conference On Recent Innovations In Science, Engineering And Technology ICRISET14,
2nd & 3rd April 2014.

102
Department of EEE & Mechanical Engg., Immanuel Arasar JJ college of Engineering