d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) e47e55

available at www.sciencedirect.com

journal homepage: www.intl.elsevierhealth.com/journals/dema

Sequential software processing of micro-XCT dental-images

for 3D-FE analysis
Flvia P. Rodrigues a, , Jianying Li b , Nick Silikas c , Rafael Y. Ballester a , David C. Watts c
a
b
c

Department of Dental Materials, School of Dentistry, University of So Paulo, So Paulo, Brazil

School of Mechanical, Aerospace and Civil Engineering, The University of Manchester, Manchester, UK
Biomaterials Science Research Group, School of Dentistry, The University of Manchester, Manchester, UK

a r t i c l e

i n f o

a b s t r a c t

Article history:

Objectives. The aim was to describe a sequential software processing of -XCT molar-images

Received 26 September 2008

for 3D-FE tooth/restoration model geometries based on a representative molar tooth, giving

Received in revised form

attention on each step of data-processing. This paper rst gives an overview of a sequential

26 January 2009

processing and then applies the resulting model to the particular case.

Accepted 2 February 2009

Methods. An intact mandibular molar was scanned using a micro-XCT instrument (1072,
SkyScan, Belgium) in which 960 slices were obtained. Sixty-three non-adjacent bitmap slices
were then optimally selected for model-creation. Enamel/dentin boundaries were claried,

Keywords:

for each slice, using image control-system software (ScanIP, Simpleware), generated a le

Micro-XCT

which was sequentially converted into a mesh in a reconstruction software (ScanFE, Simple-

FEA

ware) and posteriorly converted into a STL-le (triangulated-2D-stereolithography). This was

Dental-images

imported into a FE-software package (Patran, MSC.Software, USA) and all elements were re-

Teeth

meshed. From these elements, surfaces were created and exported to another FE-software

Occlusal loading

(Hypermesh, Altair Hyperworks) to build the dental-cavities. Finally, the volumetric-mesh

was created and the model was imported back to FE-software to apply the boundaryconditions, material-properties and initiate post-processing (using Patran and Marc, MSC
Software). To demonstrate the use of the resulting model, this was applied to the particular
case of a Class I restoration subjected to distributed loading. The analysis was performed
as linear and structural and outputs of maximum principal (MP) and maximum shear (MS)
stresses were then evaluated.
Results. A 3D-model of a mandibular molar was processed without generating errors in the
FE-package used. The maximum deviation between the tooth and the model was less than
0.1%. Stress concentrations were found at the surface where the load was applied and in
the vicinity of the toothcomposite interface.
Signicance. The described procedure is a successful method able to produce a highly detailed
3D nite element model of restored molar teeth with any cavity conguration and combination of restorative materials and this method can also be used for other biological or
biomaterials applications.
2009 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

Corresponding author at: Departamento de Materiais Dentrios, Faculdade de Odontologia, Universidade de So Paulo, Av. Prof. Lineu
Prestes, 2227, Cidade Universitria, CEP 05508-900 So Paulo, SP, Brazil. Tel.: +55 11 3091 7840x222; fax: +55 12 3951 5904.
E-mail addresses: apiro@usp.br, apiro@gmail.com (F.P. Rodrigues).
0109-5641/$ see front matter 2009 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.dental.2009.02.007

e48

1.

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) e47e55

Introduction

In recent years, there has been an increasing interest in

achieving an optimal cavity preparation design and optimal
restorative materials to minimize the stress developed by
polymerisation shrinkage and/or occlusal loading [119]. No
denitive guidelines are available yet with regard to the optimal cavity preparation design and restorative materials to be
used [18]. Studies based on laboratory experiments may give
the answers to these questions. However, a major problem is
performance on live subjects. Moreover, in view of the variability of the mechanical properties and shape of the teeth,
and since it is virtually impossible to obtain physical measurements of stresses within a tooth structure, nite element
methods (FEM) are useful tools to achieve a representation of
reality with calculation of these complex stresses. These indicate mechanical aspects of biomaterials and human tissues
that can hardly be measured in vivo [20,21]. Recently, some
authors have investigated FE optimisation methods for cavity
preparation [6,22].
FEM were introduced into dental research in 1973 [23] to
show internal stresses. Structures are divided into a number
of shaped elements with individual stress/strain characteristics [10,17]. Determination of these parameters was achieved
while the structure was exposed to external force, pressure,
thermal change, or other factors. By solving the deformation
of all the small elements simultaneously, the deformation of
the structure as a whole can be assessed [17]. On that basis
predictions can be made about failure [18,20]. Previous studies have reported different techniques to create 2D and 3D FE
models of teeth. Teeth were digitized with several types of
scanning device or even conventional computational tomography scans (CT-scan) in which the images were converted into
3D prototyping or 3D-CAD systems [24]. More simply, contours
of dental tissues were paper traced or obtained with graphics
software programs [25]. Templates were then used to create
the FE computerized model and most of them meshed with
less detail reproduction [26].
Researchers have recently shown an increased interest in
virtual FE-models and simulations associated with sophisticated techniques, in which an iterative optimization of
the model design has been performed from X-ray microcomputed tomography (XCT). Renements have been made
in geometry acquisition by recreation and digitization of planar outlines of the spatial anatomy [10,24,27]. The principle
consists in reconstructing the linear attenuation coefcient,
within an object, from the attenuation measurements of an
X-ray beam passing through the sample at different viewing
angles. Differences in linear attenuation coefcient among tissues are responsible for X-ray image contrast, which allows
quantitative analysis to be made [28].
FEM have been used to evaluate the status of restored teeth
[3,5,2934] including the tooth, interface and composite stress
and strain under simulated clinical conditions [29]. The combination of -XCT data with FE analysis result in models ner
in texture than those previously reported. By employing this
technology, the aim of this study was to describe sequential software processing of -XCT molar-images for 3D-FE
tooth/restoration model geometries based on a representative

molar tooth, giving attention on each step of data-processing.

This paper rst gives an overview of a sequential processing
and then applies the resulting model to the particular case of a Class I restoration subjected to distributed
loading.

2.

Materials and methods

The tooth (a lower second molar) was selected based on its

lack of carious lesions, and absence of abnormalities, and also
its approximate symmetry, which could simplify interpretation of results of a sound and prepared cavities tooth model
for FE analysis. Thus, the sound tooth model could serve as a
control.
To develop a 3D-FE model based upon actual geometric
dimensions, sequential software processing was performed.

2.1.
Acquisition of teeth shape data from micro-XCT
scanning
A micro-XCT was used in this study (Model 1072, Skyscan, Kontich, Belgium) for scanning the molar tooth. This instrument
is a X-ray micro-computer tomographic unit composed by the
scanner coupled to a workstation. The hardware used to process these data in the current study were a Dell Precision
workstation PWS 450 Intel Xeon (4 Gb CPU, 3.06 GHz), which
has a dedicated software for micro-XCT: SkyScan 1072 software (Skyscan, Belgium); and a Dell Precision workstation PWS
390 Intel Core (4 Gb CPU, 2.13 GHz) with TView and NRecon
(Skyscan, Belgium).
The equipment was adjusted to scan the whole tooth, with
a beam accelerating voltage of 102 kV and X-ray beam current of 96 A. Initially (using SkyScan 1072 software), a total
of 1020 Tag Image File Format (TIFF) 16-bit images, with vertical and horizontal resolution of 1336 dpi and 2 Mb size le,
were obtained. Using NRecon software, the region of interesting was then selected to generate 960 horizontal layers of
the inner structure of the tooth (transversal slices) with magnication of 14.3 pixels at 1024 1024 resolution, each as 1 Mb
bitmap les. Two frames were taken for 180 rotation per 0.45
frame angle and 4 s of exposure time per frame.
The recorded slices were then reduced to sixty-three, which
were chosen manually, according to the morphological similarities that could be noticed among les thumbnails. These
les were then organized in the same directory and checked
using TView to check if there were any areas with artefacts
and/or defects which need to be removed or non-selected
for the next step. To facilitate the visualization and workability, these bitmap were cropped with Adobe Photoshop
7.0 (Adobe Systems Incorporated, USA), or in any other software with the same tool for decreasing the le size from 1 Mb
to 280 kb and 554 508 resolution and also to focus on the
area of interesting. Un-painting also could be done if there
were still some artefacts around this area, which could add
parts that does not exist in the tooth. These sixty-three nonadjacent bitmap slices were then used for the reconstruction
process to generate the 3D-FE model-creation, with an equal
separation distance of 0.152 mm (Fig. 1), described in the next
section.

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) e47e55

e49

Fig. 1 Reconstruction process, generating 63 non-adjacent bitmap slices. The levels (from 1 to 10) were divided only for
illustrative purposes.

2.2.
Segmentation: dentin and enamel masks
creation
Enamel and dentin outer shapes, which appeared as visible
boundaries on the sixty-three bitmap slices, were accurately
claried following tooth segmentation pixel-by-pixel for each
one, using image control-system software (ScanIP , Simpleware, UK). This software imports a stack of images from -XCT
slices in a wide variety of software formats (in this case as
bitmap les), allowing steps of visualization and assisted segmentation based on image density thresholding of different
gray scale intensities corresponding to the degree of mineralization. The segmentation was then used to generate the
volumes (binary volumes) that are called masks, which dene
how the objects ll the space. This step is not fully automated
and the segmentation can be continued manually (via painting and unpainting tools, in the same way they can be done
easier in Adobe Photoshop if it is necessary). Thus, enamel
and dentin corresponding masks were created by gradually
interpolation, extraction, and lling of the borders during segmentation, changing the threshold values of two-dimensional
regions on the imported stack of images, modied until they
showed a satisfactory mask. This ScanIP le is sequentially
converted into a mesh in ScanFE (Fig. 2).

2.3.

Reconstruction for the 3D-FE model generation

The reconstruction available in NRecon is inappropriate for

FE analysis. This is normally used for checking volumes for
topological studies. To create a 3D-FE model from the step
described before, the ScanIP le with all segmented masks

was imported into ScanFE reconstruction software (Simpleware), generating a mesh of the entire image volume using
triangulated 2D shell-elements (STL-stereolithography le),
which is meshed model with information in the form required
to generate it from its underlying image and mask data. Each
mask created in ScanIP is reproduced and displayed in the
same colour within the meshed volume as a separate part in
ScanFE . The reconstruction runs in few minutes, depending
on the number of slices and masks. The STL les had a total
of 485,726 nodes and 974,464 elements, which are too large
and for use in FEA obtaining the geometry and for processing
analysis [10,35] because of their aspect ratio and connectivity
of the triangles. So these les were then imported into Patran
(MSC Software, USA) and the elements were then re-meshed to
reduce mesh density, maintaining the tooth shape. From these
new shell elements, surfaces were created (Fig. 3) and exported
to another FE software (Hypermesh , Altair Hyperworks, USA).

2.4.

Cavity design and volumetric mesh

The rst step using Hypermesh was the re-establishment

of the congruency of the dentinenamel junction, which was
lost during the previously re-meshing process. Once congruency was achieved, tooth cavities using cutting planes were
then developed (Fig. 4a). Using the same software, 3D solids
corresponded to enamel, dentin and pulp were also created
(Fig. 4b).
The nal mesh of this model consisted of four-node isoparametric tetrahedral elements (Tet4), with a mean edge
length of 0.20 mm. The number of nodal points was limited
to 37,985 associated with 201,506 elements. Finally, the vol-

e50

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) e47e55

Fig. 2 (A) Examples of segmentation steps: manual unpaint and automatic painting and interpolating using Scan IP ,
Simpleware, UK. (B) Reconstruction using Scan FE , Simpleware, UK.

umetric mesh (four-node isoparametric tetrahedral elements)

was created and the model was imported back in Patran software to apply the boundary-conditions, material-properties
and initiate FE processing (using Patran and Marc , MSC software). Tooth dimensions measured by using a digital caliper
were checked with the model ones.

2.5.
Boundary-conditions, material-properties and
analysis
To demonstrate the use of the resulting model, this was
applied to the particular case of a Class I restoration subjected to distributed loading (Fig. 5). All nodes on the lower
surface of the tooth were constrained in all directions (X,
Y and Z), preventing rigid body displacement, in agreement
with previous studies [5,9,17,36,37]. Thus it was assumed that

the overall stress distribution in the coronal portion was only

marginally affected by the root area, under the simulated
boundary-conditions [17,38]. At all interfaces, the nodes were
shared.
The occlusal distributed loading was simulated according
to precedents from the literature [3,29,39]. A uniform loading
was applied at a pressure (stress) of 10.48 MPa, corresponding
to 890 N distributed over 84.9 mm2 molar occlusal face area
[37]. The component materials were all considered homogeneous, linear, elastic and isotropic [34] and their properties
are presented in Table 1. The material colours for each group
(enamel, dentin, pulp and composite) were arbitrarily selected
from the FE software.
The analysis was performed as linear and structural. The
outputs of maximum principal (MP) and maximum shear (MS)
stresses were then evaluated.

Fig. 3 FE remeshed model and surfaces generated. (Left) The whole tooth remeshed. (Right) Surfaces of the whole tooth
(A), enamel (B), dentin (C) and pulp (D).

e51

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) e47e55

Fig. 4 (a) Tooth cavities generation using cutting planes and (b) 3D solids corresponded to enamel, dentin and pulp.

Fig. 5 Class I restoration and corresponding boundary-conditions. The model was subjected to distributed loading and all
nodes on the bottom surface were constrained in all directions (X, Y and Z).

3.

Results

A 3D model of a mandibular molar was processed without generating errors in the FE package used. The single real molar
tooth selected clearly determined the shape and geometrical dimensions of the model tooth, also allowing variations,
representing model cavity preparation. Dimensions checked
(length and diameter) from FE model were very close to those

measured by using a digital calliper, presented the geometry of the tooth very well. The maximum deviation between
the intact tooth and the corresponding model was less than
0.1%.
Characteristics stress patterns in the tooth under occlusal
force were observed. From FEM investigation on the stresses
generated by uniform loading of this tooth model, it was
evident that the stress-patterns generated in the composite material and tooth tissues were truly 3D (followed all 3D

Table 1 Elastic properties of the materials.

Material
Enamel
Dentin
Pulp
Resincomposite

Elastic modulus (GPa)

80.0
15.0
0.002
25.0

Poissons ratios
0.30
0.31
0.45
0.30

Reference
Rees and Jacobsen (1995) [12]
Ausiello et al. (2002) [3]
Barink et al. (2003) [5]
Ausiello et al. (2001) [29]

e52

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) e47e55

Fig. 6 Maximum principal stress and maximum shear stress at molar Class I cavity interface and onto the composite.

contours of the model), which suggested the potential of the

model for being versatile (Fig. 6), which can be used for simulating any structural loading (such as occlusal loading and
shrinkage).
Stress concentrations were found at the surface where the
load was applied and in the vicinity of the toothcomposite
interface, especially at the deepest and sharpest regions of
the occlusal face. Stress concentration at the dentinenamel
junction (DEJ) was also identied. Convex areas raised less
concentrated stresses than concave areas around enamel
(cavosurface angle).

4.

Discussion

It is widely recognized that FEM has proven itself as

an extremely powerful tool in addressing a wide range
of biomedical problems, mainly when complex geometries are involved [19]. Software has been developed to
assist in the generation of FE models for attending the
complexity and spatial relationship of objects and/or tissues representation. One of the possibilities is the use of
some sophisticated techniques as -XCT scanning associated

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) e47e55

with intermediate software before performing FE analysis.

Conventional scanners and prole projectors have been
used to generate 3D-models for biomedical researches related
to bone behavior, but they still required much manual work or
even resulted in coarser meshes [18,40]. Prior studies in dentistry, aiming to achieve a model from image processing, have
noticed the importance of stress distributions in a normal
tooth and the effect of different cavity designs all on the same
tooth. Some 3D-models have been prepared from contour
line teeth [41], duplicate casts [42], and dental plaster models
[29], with different cavity designs, which were sectioned and
photographed [42], laser scanned, digitized using a chargecoupled device camera attached to a stereomicroscope [26] or
CT scanned. The detailed contours of the areas of interesting
(mainly enamel, dentin and pulp chamber) were not directly
used to create a 3D-image, but traced onto millimeter-graded
papers [42] or obtained with a graphics software program
[25] with lines added to simulate different tooth preparations.
These were then digitized with a scanning device and the relative relationships of the serial longitudinal or parallel sections
to the occlusal or proximal faces were established.
Micro-XCT was developed in 1980s for in vitro studies of
bone structures [28]. By obtaining 2D slices and interpolating
the 2D information into a 3D model, this visualizes and measures complete 3D structures without sample preparation or
chemical xation [43]. The rst study in 2001 to apply this
technique to restored teeth, developed a 3D tooth-map and
exported this to FE software for stress analysis [18]. When the
generated images were directly used, the coordinate system to
stack the contours was identical to that of the scanning process and the cross-sections were automatically xed, which
facilitated few errors in the 3D mesh in comparison with other
techniques.
-XCT software is straightforward to use, but experienced
operator and good judgment will be the main key for the success of the study. There are some artifacts and noise from
-XCT data, which can be corrected by ltering when the
images are reconstructed [35]. During reconstruction, which is
a non-linear operation, any noise in the small signal areas can
produce signicant errors [43]. Hence the software incorporates tools to eliminate noise. Exposure times affect the image
contrasting and can also inuence dimensional results. With
FEM software, the difculties are directly related to the experience and knowledge of the operator and the specic problem
to be solved. In developing an FE model, some specic training
or expert collaboration is required.
The validity of a FE model refers to the possibility of it to
process with efciency, fastness and coherence all answers
front to mechanical loadings, which consequently makes it
to represent adequately the stress and strain patterns both
within the materials and their interfaces. The use of the XCT also helps generate a detailed model [18]. However, CT
data must be interpreted carefully because the accuracy could
be inuenced by the image threshold value [44]. Nevertheless,
-CT has sufcient accuracy to evaluate crown adaptation and
defects in crown restorations [20].
-XCT is contributing to many non-invasive studies
[18,24,28] but the literature is sparse on 3D-FE construction
of models and the difculties of this process. Most software

e53

has the same tools but their interfaces vary considerably in

ease of use. The main advantages of using -XCT in dentistry
are:
Relative ease of using equipment and software. Sophisticated visualization tools (shades, wireframe 3Dviews,
section views, etc.) and different lters, highlighting porosities and defects on the materials, interfaces failures, etc.
Rapid and non-destructive method [35,45]. No requirement
for staining of the objectwhich can affect the organization
of the investigated structure.
Superior resolution compared to existing digital dental systems [28]. The maximum deviation between the original CT
image and reconstructed surface solids can be less than
0.6% [19,46]; but this also depends on several factors, such
as artifacts, beam hardening, exposure time, instrument
model, etc.
3D-reconstruction from root canals [47], crowns and other
samples where the shape is more important than 3D-FE
model generation [20]; the surface areas or volumes of dental tissues can also be obtained.
Quantitative and visual measurements for biomaterials
(metals, ceramics, and polymers), including bio-histological
materials [20]; also in vitro caries research [28].
In the present study, sequential software processing of XCT dental-images to optimize 3D-FE analysis was performed.
The model was an exact replica of a second molar tooth by
the combined used of -XCT and numerical analysis software, generating a valid 3D-FE model with the possibility of
any cavity preparation, boundary-conditions and materialproperties. To demonstrate the use of the resulting model,
this was applied to the particular case of a Class I restoration subjected to distributed loading and the stress pattern
was analyzed.
The ndings corroborated [10] the advantages of using
STL les as a standard format for data transmition from
CAD to systems such as the rapid prototyping [24] and nite
element packages [10,17,19,46,48], because of their uncomplicated procedure, with minimal manual intervention. However,
to achieve congruency between these different tissues, meticulous work is still required [10].
Since teeth have many structural details or defects,
including ssures, cuspal anatomy variations, enamel anomalies, it is not realistic to incorporate too many of such details in
a model, which would lose its generality. In the present procedure, denition and biological tissue could be lost if the steps
of importing-exporting, segmentation, and meshing procedures were not done with caution.
During 2D and/or 3D mesh generation, the presence of
sharp regions makes adaptation of the elements difcult. The
software tries to deform the elements to t the geometry.
This is the rationale for using triangulated and tetrahedral
elements instead of quadrilateral ones for meshing complex
shape structures. Depending on the geometry, they may not
be meshed because of a generated error. Thus, usually during
segmentation and meshing procedures, if some areas irrelevant to the study present a non-denite shape, which does not
allow meshing, these areas can be automatically or manually
smoothed, if possible. Thus, it is at the step of segmentation,

e54

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) e47e55

when it is best to decide what is really important to simulate

in terms of dental anatomy.
For reasons of computational efciency, the anatomic
detail of the dentinenamel junction (DEJ), well captured via
the CT reconstruction, was reduced to create a smoother junction, in accordance with other studies [19]. The denition of
junctions in CT-images, such as the DEJ, must be accurate. All
junctions, between two different materials or even surfaces or
solids, sometimes need to be reconstructed manually. When
the smoothing tools are insufcient for a good result, it is benecial to work with pixel precision, not just with threshold
tools. As dental tissues are not isotropic and homogeneous,
threshold automatic tools can be non-sufcient to dene some
areas as DEJ, making them not uniform or well-smooth. In this
case, some manual work is required to guarantee the delity
of the model in this area.
The development of this 3D-FE model was part of ongoing
research into stress states of bonded restorations subjected
to polymerization shrinkage and/or occlusal loading. In this
study, only the occlusal distributed loading was simulated.
FE models predict a variety of parameters [20]; some of them
are connected with stress patterns within the structure but
does not predict failure [18] unless modeling the initiation and
propagation of cracks themselves [19]. The present results are
consistent with prior studies using 3D-FE models which used
FEM as a powerful tool for generating [10,11,18] and analyzing
the mechanical behavior of complex structures [29].
When the loading was applied, stress effects were revealed
at the cavosurface angle and DEJ. Examination of the stress
contours highlighted regions of particular interest. The stress
concentrating effect of corners is an established issue and
areas of high stress were generally expected in these regions.
Geometry acquisition and 3D modeling based on -XCT
data can still be considered at a pioneering stage [11,45,48].
Using this technique to obtain realistic 3D-FE models of teeth
and cavity preparations, further investigations are in progress
to evaluate different isthmus widths and the thickness of
residual inter-axial dentin on teeth with various cavity designs
and modes of loading. A possible limitation of this approach
may be the range of chemical elements that can be investigated, since their absorption edge must be within the X-ray
energy range of the beam line [45]. Moreover, the sequence of
software used was found workable by the authors, who do not
suggest that this is the only feasible approach to this problem.

5.

Conclusion

The purpose of the current study was to describe sequential software processing of -XCT molar-images for optimized
3D-FE tooth/restoration model geometries based on of molar
teeth, giving attention on each step of processing data. The
described procedure is a successful method able to produce a highly detailed 3D nite element model of restored
molar teeth with any cavity conguration and combination
of restorative materials and this method can also be used
for other biological or biomaterials applications. The study
develops further understanding of this new technology which
associates -XCT data and FEM.

Acknowledgements
Based in part on abstract no. 106894, presented at the 86th
IADR meeting in Toronto, July 4, 2008. The authors acknowledge Dr. Alex S. Fok for his support in the use of Scan IP
and Scan FE packages, in the School of Mechanical, Aerospace
and Civil Engineering, The University of Manchester, Manchester, United Kingdom. This study was supported in part by
FAPESP (Fundaco de Amparo Pesquisa do Estado de So
Paulo, Brazil), process no. 6/00186-3.

references

[1] Hubsch P, Middleton J, Knox J. The inuence of cavity shape

on the stresses in composite dental restorations: a nite
element study. Comput Methods Biomech Biomed Eng
2002;5:3439.
[2] Asmussen E, Munksgaard EC. Bonding of restorative resins
to dentine: status of dentine adhesives and impact on cavity
design and lling techniques. Int Dent J 1988;38:
97104.
[3] Ausiello P, Apicella A, Davidson CL. Effect of adhesive layer
properties on stress distribution in composite
restorationsa 3D nite element analysis. Dent Mater
2002;18:295303.
[4] Ausiello P, Davidson CL, Cascone P, DeGee AJ, Rengo S.
Debonding of adhesively restored deep Class II MOD
restorations after functional loading. Am J Dent
1999;12:848.
[5] Barink M, Van der Mark PC, Fennis WM, Kuijs RH, Kreulen
CM, Verdonschot N. A three-dimensional nite element
model of the polymerization process in dental restorations.
Biomaterials 2003;24:142735.
[6] Couegnat G, Fok SL, Cooper JE, Qualtrough AJ. Structural
optimization of dental restorations using the principle of
adaptive growth. Dent Mater 2006;22:312.
[7] de Trey E, Lutz F. Effect of the shape of the cavity and the
sealant system on the adaptive quality of the proximal
anterior lling margins. SSO Schweiz Monatsschr
Zahnheilkd 1977;87:73751.
[8] Feilzer AJ, De Gee AJ, Davidson CL. Setting stress in
composite resin in relation to conguration of the
restoration. J Dent Res 1987;66:16369.
[9] Hubsch PF, Middleton J, Knox J. A nite element analysis of
the stress at the restorationtooth interface, comparing
inlays and bulk llings. Biomaterials 2000;21:10159.
[10] Magne P. Efcient 3D nite element analysis of dental
restorative procedures using micro-CT data. Dent Mater
2007;23:53948.
[11] Magne P, Tan DT. Incisor compliance following operative
procedures: a rapid 3D nite element analysis using
micro-CT data. J Adhes Dent 2008;10:4956.
[12] Rees JS, Jacobsen PH. The effect of interfacial failure around
a class V composite restoration analysed by the nite
element method. J Oral Rehabil 2000;27:1116.
[13] Borkowski K, Kotousov A, Kahler B. Effect of material
properties of composite restoration on the strength of the
restorationdentine interface due to polymerization
shrinkage, thermal and occlusal loading. Med Eng Phys
2007;29:6716.
[14] Kahler B, Kotousov A, Borkowski K. Effect of material
properties on stresses at the restorationdentin interface of
composite restorations during polymerization. Dent Mater
2006;22:9427.

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) e47e55

[15] Kahler B, Kotousov A, Melkoumian N. On material choice

and fracture susceptibility of restored teeth: an asymptotic
stress analysis approach. Dent Mater 2006;22:110914.
[16] Kahler B, Kotousov A, Swain MV. On the design of dental
resin-based composites: a micromechanical approach. Acta
Biomater 2008;4:16572.
[17] Magne P, Belser UC. Rationalization of shape and related
stress distribution in posterior teeth: a nite element study
using nonlinear contact analysis. Int J Periodont Restor Dent
2002;22:42533.
[18] Verdonschot N, Fennis WM, Kuijs RH, Stolk J, Kreulen CM,
Creugers NH. Generation of 3D nite element models of
restored human teeth using micro-CT techniques. Int J
Prosthodont 2001;14:3105.
[19] Ichim I, Li Q, Loughran J, Swain MV, Kieser J. Restoration of
non-carious cervical lesions. Part I. Modelling of restorative
fracture. Dent Mater 2007;23:155361.
[20] Wakabayashi N, Ona M, Suzuki T, Igarashi Y. Nonlinear nite
element analyses: advances and challenges in dental
applications. J Dent 2008;36:46371.
[21] Fennis WM, Kuijs RH, Barink M, Kreulen CM, Verdonschot N,
Creugers NH. Can internal stresses explain the fracture
resistance of cusp-replacing composite restorations? Eur J
Oral Sci 2005;113:4438.
[22] Shi L, Fok AS, Qualtrough A. A two-stage shape optimization
process for cavity preparation. Dent Mater 2008.
[23] Farah JW, Craig RG, Sikarskie DL. Photoelastic and nite
element stress analysis of a restored axisymmetric rst
molar. J Biomech 1973;6:51120.
[24] Shimizu Y, Usui K, Araki K, Kurosaki N, Takanobu H,
Takanashi A. Study of nite element modeling from CT
images. Dent Mater J 2005;24:44755.
[25] Magne P, Douglas WH. Design optimization and evolution of
bonded ceramics for the anterior dentition: a nite-element
analysis. Quintessence Int 1999;30:66172.
[26] Magne P, Douglas WH. Interdental design of porcelain
veneers in the presence of composite llings: nite element
analysis of composite shrinkage and thermal stresses. Int J
Prosthodont 2000;13:11724.
[27] Cattaneo PM, Dalstra M, Melsen B. The nite element
method: a tool to study orthodontic tooth movement. J Dent
Res 2005;84:42833.
[28] Clementino-Luedemann TN, Kunzelmann KH. Mineral
concentration of natural human teeth by a commercial
micro-CT. Dent Mater J 2006;25:1139.
[29] Ausiello P, Apicella A, Davidson CL, Rengo S. 3D-nite
element analyses of cusp movements in a human upper
premolar, restored with adhesive resin-based composites. J
Biomech 2001;34:126977.
[30] Katona TR, Winkler MM, Huang J. Stress analysis of a
bulk-lled class-V chemical-cured dental composite
restoration. J Biomed Mater Res 1996;31:4459.
[31] Kinomoto Y, Torii M, Takeshige F, Ebisu S. Polymerization
contraction stress of resin composite restorations in a
model class I cavity conguration using photoelastic
analysis. J Esthet Dent 2000;12:30919.

e55

[32] Versluis A, Douglas WH, Cross M, Sakaguchi RL. Does an

incremental lling technique reduce polymerization
shrinkage stresses? J Dent Res 1996;75:8718.
[33] Versluis A, Tantbirojn D, Douglas WH. Distribution of
transient properties during polymerization of a
light-initiated restorative composite. Dent Mater
2004;20:54353.
[34] Versluis A, Tantbirojn D, Pintado MR, DeLong R, Douglas
WH. Residual shrinkage stress distributions in molars after
composite restoration. Dent Mater 2004;20:55464.
[35] Kato A, Ohno N. Construction of three-dimensional tooth
model by micro-computed tomography and application for
data sharing. Clin Oral Invest 2008.
[36] Ausiello P, Rengo S, Davidson CL, Watts DC. Stress
distributions in adhesively cemented ceramic and
resincomposite class II inlay restorations: a 3D-FEA study.
Dent Mater 2004;20:86272.
[37] Arola D, Galles LA, Sarubin MF. A comparison of the
mechanical behavior of posterior teeth with amalgam and
composite MOD restorations. J Dent 2001;29:6373.
[38] Goel VK, Khera SC, Ralston JL, Chang KH. Stresses at the
dentinoenamel junction of human teetha nite element
investigation. J Prosthet Dent 1991;66:4519.
[39] Anusavice KJ. Phillips: dental materials. 11th ed. Elsevier;
2005. p. 800.
[40] Akagawa Y, Sato Y, Teixeira ER, Shindoi N, Wadamoto M. A
mimic osseointegrated implant model for three-dimensional
nite element analysis. J Oral Rehabil 2003;30:415.
[41] Kubota K, Natori M, Sasaki K. Three-dimensional
measurement study of the maximum protuberance and
cervical line of the upper molars. J Nihon Univ Sch Dent
1994;36:5866.
[42] Khera SC, Goel VK, Chen RCS, Gurusami SA. A
Three-dimensional nite element model. Oper Dent
1988;13:12837.
[43] SkyScan. SkyScan 1072-Desktop X-ray
microtomograph-instruction manual; 19982001. p. 61.
[44] Kakaboura A, Rahiotis C, Watts D, Silikas N, Eliades G.
3D-marginal adaptation versus setting shrinkage in
light-cured microhybrid resin composites. Dent Mater
2007;23:2728.
[45] De Santis R, Mollica F, Prisco D, Rengo S, Ambrosio L,
Nicolais L. A 3D analysis of mechanically stressed
dentinadhesivecomposite interfaces using X-ray
micro-CT. Biomaterials 2005;26:25770.
[46] Ichim IP, Schmidlin PR, Li Q, Kieser JA, Swain MV. Restoration
of non-carious cervical lesions. Part II. Restorative material
selection to minimise fracture. Dent Mater 2007;23:15629.
[47] Cheng R, Zhou XD, Liu Z, Hu T. Development of a nite
element analysis model with curved canal and stress
analysis. J Endod 2007;33:72731.
[48] Li J, Li H, Shi L, Fok AS, Ucer C, Devlin H, et al. A
mathematical model for simulating the bone remodeling
process under mechanical stimulus. Dent Mater
2007;23:10738.