Beruflich Dokumente
Kultur Dokumente
Andreas Öchsner
Holm Altenbach Editors
Advances in
Bio-Mechanical
Systems and
Materials
Advanced Structured Materials
Volume 40
Series Editors
Andreas Öchsner
Lucas F. M. da Silva
Holm Altenbach
Advances in Bio-Mechanical
Systems and Materials
13
Editors
Andreas Öchsner Holm Altenbach
Faculty of Biosciences and Medical Lehrstuhl für Technische Mechanik
Engineering (FBME) Otto-von-Guericke-Universität
University of Technology Malaysia—UTM Magdeburg
Skudai Germany
Malaysia
and
and
Fakultät für Maschinenbau
Faculty of Engineering and Built Institut für Mechanik
Environment Magdeburg
The University of Newcastle Germany
Newcastle
Australia
The idea of this monograph is to present the latest results related to biomechanical
systems and materials. Biomechanical systems within this book are prostheses
(lower limb and orbitarian cranial cavity), implants (such as for the femoral bone,
microimplants for dental surgery, and total hip replacement), medical operation
robots (for tumor removal), and muscular retraining systems (for forearm and
wrist). To characterize and design such systems, a multidisciplinary approach is
required which involves the classical disciplines of mechanical/materials engi-
neering (design, analysis, and properties), and biology/medicine. The challenge
in such an approach is that views, concepts or even languages are sometimes dif-
ferent from discipline to discipline and the interaction and communication of the
scientists must be first developed and adjusted. In the context of materials, the
interaction of materials with mechanical systems, their description as a mechanical
system or their mechanical properties is covered.
The 6th International Conference on Advanced Computational Engineering
and Experimenting, ACE-X 2012, was held in Istanbul, Turkey, from 1 to 4 July
2012 with a strong focus on computational-based and supported engineering. This
conference served as an excellent platform for the engineering community to meet
with each other and to exchange the latest ideas. This volume contains 11 revised
and extended research articles written by experienced researchers participating in
the conference. The book will offer the state-of-the-art of tremendous advances in
biomechanical systems and materials.
The organizers and editors wish to thank all the authors for their participation
and cooperation which made this volume possible. Finally, we would like to thank
the team of Springer-Verlag, especially Dr. Christoph Baumann, for the excellent
cooperation during the preparation of this volume.
v
Contents
vii
viii Contents
1 Introduction
A. Javadzadegan · M. Behnia (*)
Department of Mechanical Engineering, The University of Sydney, Sydney, Australia
e-mail: masud.behnia@sydney.edu.au
A. Javadzadegan
e-mail: ashkan.javadzadegan@sydney.edu.au
A. S. C. Yong · L. Kritharides
Department of Cardiology, Concord Hospital, Sydney, Australia
2 Methods
2.1 Patient Cohort
FFR measurements was performed as previously published [18, 19]. EI was quan-
tified for each lesion as previously described [6].
CFD analysis was performed using ANSYS CFX (v14.0, ANSYS-Fluent Inc.,
Lebanon, NH, USA), a finite volume based software. Flow for the simulation
was assumed to be 3D, laminar and steady. The walls were considered solid and
a zero-velocity, no-slip boundary condition was adopted at the walls. Blood was
modelled as an incompressible Newtonian fluid with a dynamic viscosity of
0.0035 Pa.s and a density of 1050 kg/m3. The inlet and outlet boundary conditions
were set for the reconstructed patients’ arteries to the measured Pa and Pd.
2.5 Statistical Analysis
3 Results
Baseline clinical and lesion characteristics are shown in Table 1. 73 % of the ves-
sels assessed were the left anterior descending (LAD), 14 % were left circum-
flex (LCX) and 13 % were right coronary artery (RCA). The best cut-off value
for predicting FFR ≤ 0.75 was established by a receiver operating character-
istic (ROC) curve analysis (sensitivity of 62 %, specificity 56 %). To predict
FFR ≤ 0.75, the optimal cut-off value for EI were 0.4 (sensitivity of 62 %, speci-
ficity 56 %).
4 A. Javadzadegan et al.
Fig. 1 Relationship between fractional flow reserve (FFR) and coronary diameter stenosis
(%DS). a All lesions. b Lesions with EI ≤ 0.4. c Lesions with EI > 0.4
According to the eccentricity cut-off value, the lesions were divided into three
subgroups, overall cohort of 56 lesions, the lesions with EI > 0.4 and the lesions
with EI ≤ 0.4. Linear regression analysis demonstrated a non-significant corre-
lation between FFR and %DS (r = −0.29, P < 0.0001; Fig. 1a) and also FFR
and %AS (r = −0.37, P < 0.0001; Fig. 2a) for whole cohort of lesions. Separate
linear regression analysis revealed that there was a significant improvement in
the correlation between FFR, %DS (r = −0.62, P < 0.0001; Fig. 1c) and %AS
(r = −0.63, P < 0.0001; Fig. 2c) for the lesions with EI > 0.4, as compared to the
lesions with EI ≤ 0.4 (%DS: r = −0.12, P = 0.048; Fig. 1b, %AS: r = −0.21,
P = 0.0058; Fig. 2b).
Comparable improvement was also observed in the correlation between
FFR, MLD and MLA for more eccentric lesions (EI > 0.4) (MLD: r = 0.50,
P = 0.0001, MLA: r = 0.51, P = 0.0001), as compared with more concen-
tric lesions (EI ≤ 0.4) (MLD: r = 0.08, P = 0.11, MLA: r = 0.12, P = 0.048)
(Table 2).
1 Lesion Eccentricity and Fractional Flow Reserve and Coronary Flow 5
Fig. 2 Relationship between fractional flow reserve (FFR) and coronary area stenosis (%AS). a
All lesions. b Lesions with EI ≤ 0.4. c Lesions with EI > 0.4
4 Discussion
These data demonstrates that for overall cohort of patients, there is a weak correla-
tion between FFR and DS, MLD, MLA and AS. However, by dividing the lesion
into two subgroups; lesions with high eccentricity (above EI 0.4) and lesions with
less eccentricity (below EI 0.4), the correlations are significantly improved when
separate linear regression analysis is performed for each subgroup. This implies
that careful evaluation of lesion eccentricity during angiography may improve the
identification of functionally significant coronary stenoses.
References
1. Topol, E.J., Nissen, S.E.: Our preoccupation with coronary luminology. The dissociation
between clinical and angiographic findings in ischemic heart disease. Circulation 92, 2333–
2342 (1995)
2. Berger, A., et al.: Long-term clinical outcome after fractional flow reserve-guided percuta-
neous coronary intervention in patients with multivessel disease. J. Am. Coll. Cardiol. 46,
438–442 (2005)
3. Pijls, N.H., et al.: Percutaneous coronary intervention of functionally nonsignificant stenosis:
5-year follow-up of the DEFER Study. J. Am. Coll. Cardiol. 49, 2105–2111 (2007)
4. Potvin, J.M., et al.: Usefulness of fractional flow reserve measurements to defer revasculari-
zation in patients with stable or unstable angina pectoris, non-ST-elevation and ST-elevation
acute myocardial infarction, or atypical chest pain. Am. J. Cardiol. 98, 289–297 (2006)
5. Shaw, L.J., et al.: Optimal medical therapy with or without percutaneous coronary
intervention to reduce ischemic burden: results from the Clinical Outcomes Utilizing
Revascularization and Aggressive Drug Evaluation (COURAGE) trial nuclear substudy.
Circulation 117(10), 1283–1291 (2008)
6 A. Javadzadegan et al.
6. Li, F., et al.: The association of lesion eccentricity with plaque morphology and components
in the superficial femoral artery: a high-spatial-resolution, multi-contrast weighted CMR
study. J. Cardiovasc. Magn. Reson. 12, 37 (2010)
7. Yamagishi, M., et al.: Morphology of vulnerable coronary plaque: insights from follow-up
of patients examined by intravascular ultrasound before an acute coronary syndrome. J. Am.
Coll. Cardiol. 35(1), 106–111 (2000)
8. Brosh, D., et al.: Effect of lesion length on fractional flow reserve in intermediate coronary
lesions. Am. Heart J. 150, 338–343 (2004)
9. Kolozsvari, R., et al.: Plaque volume derived from three-dimensional reconstruction
of coronary angiography predicts the fractional flow reserve. Int. J. Cardiol. (2011).
doi:10.1016/j.ijcard.2011.04.010
10. Briguori, C., et al.: Intravascular ultrasound criteria for the assessment of the functional
significance of intermediate coronary artery stenoses and comparison with fractional flow
reserve. Am. J. Cardiol. 87, 136–141 (2001)
11. Jasti, V., et al.: Correlations between fractional flow reserve and intravascular ultrasound in
patients with an ambiguous left main coronary artery stenosis. Circulation 110, 2831–2836
(2004)
12. Takagi, A., et al.: Clinical potential of intravascular ultrasound for physiological assessment
of coronary stenosis: relationship between quantitative ultrasound tomography and pressure-
derived fractional flow reserve. Circulation 100, 250–255 (1999)
13. Tobis, J., et al.: Assessment of intermediate severity coronary lesions in the catheterization
laboratory. J. Am. Coll. Cardiol. 49, 839–848 (2007)
14. Lamm, C., et al.: High-fidelity translesional pressure gradients during percutaneous translu-
minal coronary angioplasty: correlation with quantitative coronary angiography. Am. Heart J.
126, 66–75 (1993)
15. Hanekamp, C.E.E., et al.: Comparison of quantitative coronary angiography, intravascu-
lar ultrasound, and coronary pressure measurement to assess optimum stent deployment.
Circulation 99, 1015–1021 (1999)
16. Abizaid, A., et al.: Clinical, intravascular ultrasound and quantitative angiographic determi-
nants of the coronary flow reserve before and after percutaneous transluminal coronary angi-
oplasty. Am. J. Cardiol. 82, 423–428 (1998)
17. Moses, J.W., et al.: Relation between single tomographic intravascular ultrasound image
parameters and intracoronary Doppler flow velocity in patients with intermediately severe
coronary stenoses. Am. Heart J. 135, 988–994 (1998)
18. Yong, A.S.C., et al.: Three-dimensional and two-dimensional quantitative coronary angiog-
raphy, and their prediction of reduced fractional flow reserve. Eur. Heart J. 32(3), 345–353
(2011)
19. Javadzadegan, et al.: Flow recirculation zone length and shear rate are differentially affected
by stenosis severity in human coronary arteries. Am. J. Physiol. Heart Circ. Physiol. 304,
559–566 (2013)
20. Ng, M.K., et al.: Invasive assessment of the coronary microcirculation: superior reproducibil-
ity and less hemodynamic dependence of index of microcirculatory resistance compared with
coronary flow reserve. Circulation 113(17), 2054–2061 (2006)
Design of a Biomechatronics Robot
to Provide Therapy and to Remove Tumors
tumors, with the purpose of contributing in the design of equipment that serves
to make the manual labor that perform an oncologist with patients in Chihuahua,
México.
Abbreviations
1 Introduction
The results obtained so far in building the robot consists of two parts: (a) the
designs: solid works, mechanical, electronic and computational; and (b) modeling
and simulation.
Talking about the medical functions that the robot is supposed to perform, it
must execute the work of radiosurgery, as the medical procedure of radiotherapy
is called, where narrow beams of radiation in the megavoltage are administered by
multiple convergent and formed fields. This allows having a high dose of irradia-
tion which can be accurately located in an area or specific anatomical structure,
avoiding toxic doses to adjacent tissues [12–16].
The terminals are expected to provide therapy to remove damaged tissue
tumors, and to perform a precise function through the correct location of the
affected tissue area as in radiosurgery. To do this, we modeled and simulated
the functions of human tissue with the Comsol Multiphysics software [17–22].
The temperatures are expected to warm the terminals and the damaged tissue
where the therapy and/or removal is undergoing, and reach the ideal tempera-
tures to perform the medical tasks.
Despite all the wealth and new technologies in robot developments, there is no
Cartesian robot that performs the heat transfer function with applications in oncol-
ogy [12–16] at present in the market.
The first tests will be conducted with animal tissue during the II/2012 semester,
and precision and refinement testing of the functions for which the robot was
designed, is expected to be completed in the I/2013 semester with the conclusion
of the student`s theses1 that are working on this project.
The Solid Works design of the pieces of the robot is original, and all parts were
machined with nylacero which is a modified copolymer lauryl lactam with greater
tensile strength and impact resistance. This type of Cartesian robots tackles a lot of
automation tasks such as handling and assembly of parts and performing reliably,
quickly and cost-effective of several functions. This offers the enormous poten-
tial to use the same robot for various applications, thanks to the modular robotic
systems.
The Solid Works 2009 design [10] for the robot without medical terminals has
the copyright registration No. 03-2010-012012301500-01 and is entitled:
Cartesian robot with three degrees of freedom.2
The software was chosen to simulate the mechanical design which is an eas-
ily readable model. Furthermore, it allows the simulation and optimization of the
parameters using only standard values, such as number of teeth, diametral pitches,
materials, among others, and presents results in a graphical and numerical form
1 Scholarships DGEST: 03 2010-068, 069, 070, 071 and 072 ID of thesis project for: D. J.
Camacho Medina, J. Ochoa Morales, O. G. Ochoa González, A. Ruiz Mendoza and D. A.
Villalba Caraveo, respectively. Moreover, J. I. Loya Valladares, L. A. Rodríguez Gándara,
W. Ceballos Castillo, A. E. Ledezma Lechuga, L. E. Torres González, I. Arias Chávez, D. A.
Rodríguez García, M. A. Torres Navarro.
2 Degree Theses of Emmanuel Núñez Jáquez.
10 R. Leticia Corral-Bustamante et al.
and has substantial improvements over previous versions. Solid Works is a pro-
gram for 3D mechanical design that uses a graphical environment based on
Microsoft Windows, intuitive and easy to handle. The main features that make
Solid Works a versatile and accurate tool is its ability to create assemblies from
parts, the fact that the parts can be easily modified and uses standard measures
in a bidirectional way with all applications. Also it uses the layout manager
(Feature Manager) that facilitates the rapid change in three-dimensional opera-
tions sketch without having to redo the design already reflected in the rest of their
associated documents. Along with the part design tools, assemblies and drawings,
SolidWorks [10] includes productivity tools, project management, presentation,
analysis and simulation in its mechanical design.
2 Nomenclature
3 Modeling
Modeling of the robot functions was performed by the Comsol Multiphysics soft-
ware [17–22], which makes use of the heat transfer capacity [3–9] to simulate
therapy and the removal of damaged tissue cells. The human tissue is modeled as
a cylindrical geometry to which a conductive heat is applied to burn the cancer
cells. The module of heat transfer in biological tissues uses the bioheat transfer
interface of Comsol and plays an important role in technology for medical pur-
poses. Comsol uses the approximation of Penne [4] to represent the heat sources
from metabolism and blood perfusion. The equation for heat transfer by conduc-
tion with this approach is
∂T
ρCp + ∇ · (−κ∇T ) = ρb Cb ωb (Tb − T ) + Qmet (1a)
∂t
The terms on the left-hand side of Eq. (1a) belong to the model of biological tissue,
while the terms on theright-hand side provide; provide the bioheat model [3].
Tumor ablation involves passing of four electrodes at a given temperature
through the affected tissue. The method involves inserting a tube in which elec-
trical current flows through four electrodes leaving a plunger, to the well local-
ized cancerous tissue. Radio frequencies are used to heat these electrodes due to
heat transfer [3–9] up to a temperature between 45 and 50 °C in the tissue. This
method serves to increase the cell temperature above 45–50 °C, resulting in pro-
tein denaturation with coagulation that is the ultimate cause of cell death and
tissue necrosis. RF tumor ablation could be implemented in patients with liver
tumors [23, 24], kidney, lung [25], prostate and breast, among others. Currently, a
radiation oncologist [12–16, 25] performs this function by hand as far as informa-
tion is available in the State of Chihuahua.
The probe is a needle (main bar) and four electrode arms as shown in Fig. 1.
The needle is electrically isolated, except near the electrode arms. An electric cur-
rent through the probe creates an electric field in the tissue. The field is the strong-
est in the immediate vicinity of the probe and generates resistive heating, which
dominates around the arms of the probe electrode due to the strong electric field.
This model uses the bioheat equation and the continuous current mode to
implement a transient analysis
∂T
δts ρC + ∇ · (−κ∇T ) = ρb Cb ωb (Tb − T ) + Qmet + Qext (1b)
∂t
where,
∂ φ̂ ∂ ∂
∇ = ρ̂ + + ẑ
∂t ρ ∂φ ∂z
and,
J e = Qj = 0
−∇ · σ ∇V − Je = Qj −→ −∇ · (σ ∇V ) = 0 (2)
12 R. Leticia Corral-Bustamante et al.
(b)
1.79 mm
Conductor
Dielectric
Catheter
(a)
1 mm 5 mm
Fig. 2 Antenna geometry for microwave coagulation therapy. a A coaxial cable with a ring-
shaped slot cut in the outer conductor which produces a short circuit at the tip. b A plastic cath-
eter surrounding the antenna
It is assumed in the computational model that the body tissue is a cylinder and
its temperature remains at 37 °C during the entire process. The model locates the
probe along the centerline of the tissue such that its electrodes cross the region
where the tumor is located.
The terminal to perform the cancer therapy by microwave consists of the fol-
lowing elements (see Fig. 2a): an instrument that converts electrical energy to
electromagnetic energy by a coaxial antenna that emits a microwave.
Design of a Biomechatronics Robot to Provide Therapy 13
The main instrument is a physical device that carries electrical energy to a very
small antenna which emits microwaves onto a piece of human tissue.
The top of the antenna with a load of 128–1,300 milli volts is introduced to the
center of the tumor, emitting microwaves, creating friction in the tissue of the mol-
ecules, generating heat transfer (electromagnetic field modeled by Eq. (1a, b) to a
temperature of 50 °C to kill cancer cells.
The innovation that is intended to provide is to control three fundamental
aspects: the induced voltage and time per session controlled by microcontrollers
and temperature in order to achieve a good therapy.
The modeling of the microwave cancer therapy is done in two-dimensions with
cylindrical coordinates as the problem is of rotational symmetry (see Fig. 2b).
An electromagnetic wave propagating in a coaxial cable is characterized by
transverse electromagnetic fields (TEM). Assuming harmonic time fields with com-
plex amplitudes containing the phase information, the appropriate equations are:
C j(ωt−κz)
E = er e (3)
R
C j(ωt−κz)
H = eϕ e (4)
Z
router
1
C2
router
Pav = Re E × H∗ 2π rdr = ez ln (5)
rinner 2 Z rinner
to an input power of W Pav as shown in the average power flow in time. The
antenna radiates in the tissue where a damped wave propagates: (1) an absorption
boundary condition at some distance from the antenna, without excitement, in all
the external borders, and (2) A symmetry boundary condition at the boundaries
r = 0, namely
Er = 0 (11)
∂Ez
=0 (12)
∂r
The domain and boundary equations of heat transfer are satisfied by the bioheat
equation which describes this phenomenon in steady state as
∇ · (−κ∇T ) = ρb Cb ωb (Tb − T ) + Qmet + Qext (13)
Qmet = 0 (14)
1
Re (σ − jωε) E · E∗ (15)
Qext =
2
The model assumes that the rate of perfusion of blood is ωb = 0.0036 s−1, and
that the blood enters the liver at a body temperature of Tb = (37 + 273.15) K and
is heated up to a temperature T. The heat capacity of the blood is Cb = 3,639 J/
kg K. These data are used to model the heat transfer only in the domain of the
liver. Where this field is truncated, the insulation is used, namely:
n · ∇T = 0 (16)
Fig. 4 Electronic design and analysis of shape memory material: (a) the temperature controller
circuit, (b) printed circuit board, (c) soldier components, (d) full box (Push button, indicators,
LCD displays, switch and connection to electrodes) and (e) terminals and box components (ter-
minal stripping of liver tumors: shortest amount of time and high accuracy to improve security of
the process)
2. Control circuit.
Receives signals from the majority of the circuits.
Controls output to the electrodes.
Gives way to the RF to the electrodes when the start button is pressed and it
does not emit this signal when the device is not operating or when the process
ends. This signal is received from the timing circuit.
Receive signals from the circuit of the electrode to maintain its temperature at
55 and 65 °C.
Control the input and output or the electrodes on the needle.
Design of a Biomechatronics Robot to Provide Therapy 17
Fig. 5 Block diagram of the complete circuit for electronically controlling the function of the
terminals of the robot
Fig. 6 Block control-
frequency. a RF transmitter
circuit. b Output signal RF
circuit
7. The circuit for controlling the temperature of the terminal to provide cancer
therapy by microwave emission.
With respect to the computational design, the micro controller programming
was performed to play its role in the mechatronic prototype.
The control of the circuit through the micro controller (3 pieces) PIC16F873A,
each was charged by separate programs according to their function in the circuit.
Program for function 1: Controls the duration of the process, sends signals to
insert and remove electrodes and emits signals to start or stop the process. For the
program for function 2 there is a PIC that indicates the status of the process as
waiting, treatment in process and treatment completed. Finally, the program for
function 3 has a PIC to control the temperature (an input (sensor LM35) and an
output).
The RC3GLcontrol software was developed in three programming lan-
guages of high and intermediate level: Flash CS5, Action Script 2.0 (Design of
Design of a Biomechatronics Robot to Provide Therapy 19
Fig. 7 a Physical circuit of temperature. b Circuit of time control. c Circuit of state of process. d
All circuit in protoboard
the appearance of software in Flash CS5 using Action Script 2.0 code) and Visual
Basic.NET. A video test was performed in the control software.
The simulation of the control of the stepper motor in Proteus 7.7SP2 with a
PIC16 series microcontroller and the simulation in Proteus 7.7SP2 of the interac-
tion between the PIC18F4550 and USB port of the computer and communication
20 R. Leticia Corral-Bustamante et al.
Code in Graphical
Software in Action Environment
Visual Basic Script 2.0
In Flash CS5
User interac-
tion with the USB port PIC18F4550
software
Position
PIC16F877 sensors, lim- Video devices
it switches
Precise
Driver Stepper
movements
ULN2803
motors of the robot
with the control software: Circuit of control and power to establish communication
and control the movements of the robot.
Table 1 shows the process for optimal control by means of circuit of control
and power of the RC3GL used to test with two of the four stepper motors.
The results we expect to obtain during the implementation of the mechatronic
robot have been modeled only theoretically in the software Comsol Multiphysics
[17–22] and the graphics are contained in its Model Library for: (a) removal of
tumor and (b) microwave cancer therapy.
During the process of tumor removal, a temperature field in the intervals of
39–82.017 °C and/or 310 and 361 K is expected in the electrodes during a period
of 60 s. It is also expected that the temperature at the tip of an electrode behaves as
the graph of Fig. 8.
A localized region of human tissue must reach 50 °C within 8 min and must be
swept by the electrodes [23].
During the process of cancer therapy by means of microwaves (for steady state
and a microwave power input of 10 W), it is expected that the temperature field in
Design of a Biomechatronics Robot to Provide Therapy 21
Fig. 8 Expected
temperatures at the tip of an
electrode
Temperature [ºC]
Time [s]
The blood perfusion which is relatively cold seems to limit the extension of the
area that is being heated.
Close to the antenna, the heat source is stronger (see Fig. 2), and far from it,
the heat source is weaker and the blood manages to keep the tissue at normal body
temperature.
The normalized specific absorption (SAR) values along a line parallel to the
antenna should show a behavior as the graph of Fig. 9.
5 Conclusions
References
1. Zsuzsanna, M.L., McDannold, N.J., Fennessy, F.M., et al.: Uterine leiomyomas: mr imaging–
guided focused ultrasound surgery—imaging predictors of success. Radiol (2008). doi:10.11
48/radiol.2491071600
2. Sandoval-Rodriguez, R., Abdallah, Ch., Jerez, H., et al.: Networked control systems: algo-
rithms and experiments. In: Chiasson, J., Loiseau, J.J. (eds.) Applications of Time Delay
Systems, pp. 37–56. Springer, Heidelberg (2007)
3. Corral Bustamante, R.L.: Termofluidos: fundamentos de termodinámica y análisis de fluidos.
Editorial Académica Española, Lap Lambert Academic Publishing GmbH & Co, KG (2012)
4. Bejan, A.: Heat Transfer. Wiley, New York (1993)
5. Cengel, Y.A., Turner, R.H.: Fundamentals of Thermal-Fluid Sciences. Mcgraw-Hill, New
York (2007)
6. Horsley, M.: Thermofluids. Taylor & Francis, London (1996)
7. Incropera, F.P., DeWitt, D.P.: Fundamentos De Transferencia De Calor. Prentice Hall,
México (1999)
8. Kaminski, D.A., Jensen, M.K.: Introduction to Thermal and Fluid Engineering. Wiley, USA
(2005)
9. Myers, G.E.: Analytical Methods in Conduction Heat Transfer. AMCHT Publications, USA
(1998)
Design of a Biomechatronics Robot to Provide Therapy 23
10. Gómez, S.: El gran libro de Solidworks Office Professional. Alfaomega, Bogotá (2008)
11. Corral Bustamante, R.L., Núñez Jáquez, E.: Robot Cartesiano de 3 Grados de Libertad.
Indautor, México (2010)
12. González,D.L.H., M et al.: Radioterapia. Mini Manuales Prácticos: Oncología. Arán
Ediciones, Madrid (2008)
13. Halperin, E.C., Perez, C.A., Brady, L.W.: Principles and Practice of Radiation Oncology.
Lippincott Williams & Wilkins, Philadelphia (2008)
14. Kha, F.M.: The Physics of Radiation Therapy. Lippincott Williams & Wilkins, Philadelphia
(2010)
15. Lawrence, T.S., Haken, R.K.T., Giaccia, A.: Principles of radiation oncology. In: DeVita
V.T. Jr, Lawrence, T.S., Rosenberg, S.A. (eds.) DeVita, Hellman, and Rosenberg’s Cancer:
Principles and Practice of Oncology, pp. 307–336. Lippincott Williams & Wilkins,
Philadelphia (2008)
16. Martins, Y., Lederman, R.I., Lowenstein, C.L., et al.: Increasing response rates from physi-
cians in oncology research: a structured literature review and data from a recent physician
survey. Br. J. Cancer. doi (2012). doi:10.1038/bjc.2012.28
17. Zimmerman, W.B.J.: Multiphysics Modeling with Finite Element Methods. World Scientific
Publishing, Singapore (2006)
18. Torres, R., Grau, J.: Introducción a la Mecánica de Fluidos y Transferencia de Calor Con
COMSOL Multiphysics. Addlink Media, España (2007)
19. Pryor, R.W.: Multiphysics modeling using COMSOL® v.4. Mercury Learning and
Information, David Pallai, Canada (2012)
20. Battaglia, J.L.: Heat Transfer in Materials Forming Processes. Wiley-ISTE, USA (2008)
21. Datta, A., Rakesh, V.: An Introduction to Modeling of Transport Processes: Applications to
Biomedical Systems. Cambridge University Press, New York (2010)
22. Plawsky, J.:Transport Phenomena Fundamentals. Marcel Dekker, New York (2001)
23. Tungjitkusolmun, S., Staelin, S.T., Haemmerich, D. et al: Three-dimensional finite element
analyses for radio-frequency hepatic tumor ablation. IEEE Trans. on Biomed. Eng. 49(1),
3–9 (2002). http://www.rf-ablation.engr.wisc.edu/papers/Supan_Hepatic.pdf
24. Shen, B., Chu, E.S.H., Zhao, G., et al.: PPARgamma inhibits hepatocellular carcinoma
metastasesin vitro and in mice. Br. J. Cancer 106, 1486–1494 (2012)
25. van der Voort van Zyp, N.C., Prévost, J.B., van der Holt, B.: Quality of life after stereotactic
radiotherapy for stage I non-small-cell lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 77(1),
31–7 (2010)
26. López-Longás, J., Canut, J., Ríos, R., et al.: Caracterización termo-mecánica de la aleación
Ni-Ti para aplicaciones en medicina. Bioméc. 6(11), 73–80 (1998)
27. Saito, K., Taniguchi, T., Yoshimura, H. et al.: Estimation of SAR distribution of a tip-split
array applicator for microwave coagulation therapy using the finite element method. IEICE
Trans. Electron. E84-C(7), 948–954 (2001)
Nanostructured Hydroxyapatite Coating
for Biodegradability Improvement of
Magnesium-based Alloy Implant
1 Introduction
Many efforts have been implemented to produce an implant for injured bones
with the main following traits: biocompatibility to avoid toxicity of the tissues [1],
good mechanical properties for load bearing [2], and biodegradability for eliminat-
ing second surgery procedure for removing plates, bone screws and other fixture
devices to assist speeding up the tissue amendment [3]. Magnesium alloys have
met most of the requirements. Results show that magnesium has admissible bio-
compatibility and causes cell proliferation [4]. Also, they have good mechanical
properties and their elastic modulus is close to that of the natural bones which
avoid stress shielding effects [2]. Stress shielding is a mechanical concept that
could be described as follows; when two or more components with different elas-
tic moduli make up a mechanical system, redistribution of load will occur among
them. The component with a higher elastic modulus will bear more loads and the
component with a lower elastic modulus will tolerate fewer loads [5]. This phe-
nomenon threatens the integrity of the bone restoration process [6] and should be
avoided in bone supporters designing. Noteworthy that magnesium is one of the
most essential elements for metabolic activities. It has been found that the dose of
magnesium required for human body is about 300–400 mg per day [7].
The greatest issue in utilization of magnesium for bone implants is its high cor-
rosion rate in physiological solutions due to presence of chloride ions [8].
Alloying magnesium with alloying elements which are nobler than magnesium
such as aluminum, zinc, rare earth elements [9], manufacturing magnesium—
bioceramics composites [10], and surface modification techniques [11] are some
of the routes to control the magnesium degradation rate. The aim of this work was
to improve biodegradation of magnesium based alloys by applying the n-HA coat-
ing via sol–gel method. Former investigations showed that sol–gel primary particles
products have nano-sized dimension [12]. Bose et al. showed that nanocrystalline
hydroxyapatite will increase bone cell adhesion and proliferation against micrometer
range hydroxyapatite [13].
Nanostructured hydroxyapatite includes the main mineral phase of hard tissue in
biologically produced organic matrix (approximately 60 %) [14]. It is a good candidate
to be applied as an implant coating to reduce the substrate degradation rate. Besides,
a strong chemical bonding will occur at the interface of the tissue and hydroxyapatite
coated implant after implantation, which will increase the tissue regeneration rate [15].
2.2 Characterization
The coating has been scratched from the surface of the coated specimen after the sin-
tering process. X-ray diffraction technique (XRD, Philips Xpert) with a Cu Kα wave-
length of 1.5418 angstrom (Å) was utilized to analyze the crystal structure, phase
composition and crystallinity of the prepared n-HA, and confirm the AZ91 structure.
The diffractometer was operated at 40 kV and 30 mA employing a step size of 0.02˚/s.
The fraction of the crystalline phase (Xc) in the hydroxyapatite nanopowders
can be evaluated by Eq. (1).
Xc = 1 − V112/300 /I300 (1)
where I300 is the intensity of (300) diffraction peak and V112/300 is the intensity of
the hollow between (112) and (300) diffraction peaks of hydroxyapatite [16].
The peak broadening of X-ray diffraction pattern can be used to estimate the
crystallite size in a direction perpendicular to the crystallographic plane based on
Scherer’s formula as Eq. (2) [16].
Xs = 0.9λ/FWHM.cosθ (2)
where Xs, λ, FWHM, Θ are the crystallite size (nm), the wavelength of X-ray
beam (λ = 1.5418 Å for Cu Kα radiation), full width at half maximum for the dif-
fraction peak under consideration (radian), and Θ is the diffraction angle (degree).
0.5 wt % of n-HA powdered sample (scratched from the surface of the coat-
ing) was mixed with 99.5 wt % of KBr as an infrared grade and pelletized under
28 R. Rojaee et al.
The bare and n-HA coated AZ91 specimens with the effective area of 1 cm2 were
incubated in 10 ml of simulated body fluid (c-SBF2) in gamma-sterilized 50 ml fal-
con bottles by the Bohner and Lemaitre protocol [17]. c-SBF2 solution is the revised
simulated body fluid (c-SBF) which was proposed by Kokubo and Takadama [18].
Afterwards, the falcons were placed in a water bath at 37 ± 1°C within 28 days. The
volume of simulated body fluid used for testing determined by Eq. (3).
Vs = Sa /10 (3)
where Vs is the volume of SBF (ml), and Sa is the apparent surface area of the
specimen (mm2) [18].
In order to determine degradation rate, various groups of specimens were
immersed in SBF. The specimen’s weight was measured before and after the
immersion by a balance with an accuracy of 10−5 g. The immersed specimens
were cleaned using 200 g/l chromic acid to remove the surface corrosion product,
rinsed with double distilled water, dried in air and finally weighted to calculate the
weight loss. The corrosion rate was calculated using Eq. (4).
CR = W /A.t (4)
where CR is the corrosion rate, W is the weight loss, A is the surface area exposed
to the corrosive media and t is the exposure time.
Magnesium ion concentrations of SBF solution, before and after soaking the
specimens, were measured using inductively coupled plasma-optical emission
spectrometry (ICP-OES; OPTIMA 7300 DV).
In order to determine the hydrogen evolution rate, specimens were immersed in
synthesized simulated body fluid in a glassware setup. The schematic of the setup is
shown in Fig. 1. The variation of liquid’s height had been recorded within 8 weeks.
Fig. 1 Schematic illustration
of hydrogen evolution setup
system. The counter electrode was a standard platinum electrode and the reference
electrode was a standard Ag/AgCl electrode. In order to statistical study the corro-
sion behavior; various groups of uncoated and coated specimens have been used
for potentiodynamic polarization tests. The electrochemical corrosion curves in
this chapter were selected for presentation because their extracted data were the
nearest to the mean values of the corrosion current densities and mean values of
the corrosion potentials of each group of specimens. Due to potential-time curves
in Fig. 2, the specimens maintained one hour in SBF electrolyte in the standby
condition before commencing potentiodynamic polarization tests. This decision
was made because a steady open-circuit potential (not more than ± 5 mV drift
in 5 min) was attained at about one hour soaking the specimens in the electro-
lyte [19]. The electrochemically concept of the steady state condition is forming
and stabilizing of electrolyte ions layer and negatively charged surface of a metal
around a soaked specimen which is called electrical double layer (EDL) [20].
The corrosion current densities and corrosion potentials of two groups of speci-
mens were determined from the potentiodynamic polarization curves by linear polar-
ization and Tafel extrapolation methods by Ametek potentiostat (model PARSTAT
2273) at 37 ± 1°C at a potential scanning rate of 1 mV/s initiated at −250 mV
below the open circuit potential and the atmosphere was open to air. The mean value
and standard deviations of the results were also calculated. As open circuit potential
versus elapsed time graph shows in Fig. 2, for both of bare and n-HA coated AZ91
30 R. Rojaee et al.
Fig. 2 Open circuit potential versus elapsed time for sol–gel n-HA coated AZ91 (curve # 1), and
uncoated AZ91 (curve # 2) in SBF solution electrolyte at 37 ± 1°C
specimens, the electrode potential tends to become nobler with time. Also, a shift of
potential towards positive direction is observed for the coated specimen.
3.1 Characterization
The FT-IR spectrum of a n-HA sample is shown in Fig. 3. All peaks of pure syn-
thesized hydroxyapatite can be observed in the FT-IR spectra. As portrayed in
Fig. 3, broad FT-IR υ3 (bending mode) P–O band at about 1,000–1,100 cm−1
is the first indication shows that hydroxyapatite is formed. Also υ1 (symmetric
stretching mode) was observed in 962.30 cm−1 in FT-IR spectra, and υ4 (bend-
ing mode) vibration of the PO43− ion was detected in 571.79, and 600.72 cm−1.
The bands observed at 1624.73 and 3432.67 cm−1 indicate adsorbed H2O in
the examined powder. A weak band at 3571.52 cm−1 is due to structural O–H
which appears as shoulders. Another obscure structural O–H bending band in
hydroxyapatite occurs at 630.61 cm−1 [21]. The FT-IR absorption bands confirms
the presence of carbonate at 1421.28, and 1457.92 cm−1 which can be because of
atmosphere and incomplete pyrolysis of organics precursors which may have dis-
solved into the hydroxyapatite crystal [22, 23].
Nanostructured Hydroxyapatite Coating for Biodegradability Improvement 31
X-ray diffraction pattern of the AZ91 specimen is shown in Fig. 4. According to Fig. 4,
the AZ91 microstructure is composed of α-phase (pure Mg), and β-phase (Al12Mg17).
The X-ray diffraction pattern of synthesized hydroxyapatite is shown in Fig. 5. It has
been found that there is a good match with the synthesized and standard hydroxyapatite
with the PDF no. 9-432 both in terms of intensity and position of the peaks.
Crystallinity of sol–gel synthesized hydroxyapatite nanopowders which has been
sintered at 400°C is about 45 % according to Eq. (1). Otherwise, more crystallinity
of hydroxyapatite nanopowders would be occurred by increasing the sintering tem-
perature [16], low melting temperature of AZ91, and also its high tendency to react
with oxygen by increasing temperature especially above 400°C, restrict the n-HA
coated AZ91 specimens to encounter more than 400°C sintering temperature [24].
The estimation of synthesized hydroxyapatite nanopowders crystallite size has been
performed according to Eq. (2) by choosing the (002) peak. This peak was selected
since it was sharper and isolated from others. The estimated crystallite size of syn-
thesized hydroxyapatite nanopowders is calculated to be around 40 nm. Table 2
gives the diffraction plane, d spacing, and average size data.
Figure 6b, c depict the SEM micrographs of the surface of the coating in differ-
ent magnifications. It is obvious that a crack-free and homogeneous n-HA coating
has been obtained on AZ91 substrate via sol–gel method.
Nanostructured Hydroxyapatite Coating for Biodegradability Improvement 33
Table 2 Size, diffraction plane, d spacing of hydroxyapatite powder. These data have been
extracted from hydroxyapatite X-ray diffraction pattern in Fig. 5
Diffraction angle FWHM (radians) d spacing (nm) Diffraction plane Size (nm)
2 Theta (degree)
25.921 0.0041 3.4374 (002) ≈ 40
Fig. 6 a TEM image of n-HA powder obtained from the surface of coating. SEM micrographs
of the n-HA coating on AZ91 in b low magnification, c high magnification. SEM micrographs
of the precipitated apatites after 28 days of immersion in SBF solution at 37 ± 1°C on d Bare
AZ91, e n-HA coated AZ91
Figure 6d, e were monitored after 28 days immersion of both bare and n-HA
coated AZ91 specimens in SBF solution. Some big pits have been revealed in the
bare AZ91 specimen (Fig. 6d) which indicates its corrosion mechanism in SBF
34 R. Rojaee et al.
solution. It should be noted that apatite-like sediments are formed on both types of
specimens but more density of apatite-like sediments are precipitated on the sur-
face of n-HA coated specimen, as expected.
Fig. 7 Volume of hydrogen evolved for bare and n-HA coated AZ91 in SBF at 37 ± 1°C
Nanostructured Hydroxyapatite Coating for Biodegradability Improvement 35
Mg2+ −
(aq) + 2OH(aq) ⇆ Mg (OH)2 (s) (product formation) (7)
Fig. 8 Potentiodynamic polarization curves of sol–gel n-HA coated AZ91 (curve # 1), and
uncoated AZ91 (curve # 2) in SBF solution electrolyte at 37 ± 1 °C
36 R. Rojaee et al.
Table 3 Mean values (standard deviation) of corrosion current densities and corrosion potentials
in SBF at 37 ± 1°C
Material Ecorr. Vs. Ag/AgCl (mV) ICorr. (μA/cm2)
Tafel Linear Tafel Linear
Bare AZ91 −1448 (51) −1427 (32) 22.14 (3.21) 14.12 (1.32)
n-HA coated AZ91 −1343 (35) −1296 (26) 2.83 (0.04) 2.07 (0.03)
Fig. 9 pH variations of SBF immersed with n-HA coated AZ91 and bare AZ91 within 28 days
3.4 Bio-Degradation Evaluation
Figure 10 shows that the corrosion rate decreases as the immersion time increases.
n-HA coated AZ91 specimens degraded more slowly than bare AZ91 ones, which
is in good agreement with the electrochemical results in Sect. 3.3. The reason is
that both bare and n-HA coated AZ91 specimens were covered with partially pro-
tective corrosion products and apatite precipitations on their surface, which lower
their degradation rates [29].
3.5 ICP-OES Analysis
Fig. 10 Variation of the degradation rate of n-HA coated AZ91 and bare AZ91 specimens in
SBF solution as a function of immersion time at 37 ± 1°C
38 R. Rojaee et al.
Fig. 11 Variation of Mg2+ ions concentrations in SBF solution immersed with bare AZ91 and
n-HA coated AZ91 within 28 days
4 Conclusion
References
1. Geng, F., Tan, L.L., Jin, X.X., Yang, J.Y., Yang, K.: The preparation, cytocompatibility, and
in vitro biodegradation study of pure beta-TCP on magnesium. J. Mater. Sci. Mater. Med. 20,
1149–1157 (2009)
2. Yan, T., Tan, L., Xiong, D., Liu, X., Zhang, B., Yang, K.: Fluoride treatment and in vitro cor-
rosion behavior of an AZ31B magnesium alloy. Mater. Sci. Eng. C. 30, 740–748 (2010)
3. Razavi, M., Fathi, M.H., Meratian, M.: Microstructure, mechanical properties and bio-cor-
rosion evaluation of biodegradable AZ91-FA nanocomposites for biomedical applications.
Mater. Sci. Eng. A. 527, 6938–6944 (2010)
Nanostructured Hydroxyapatite Coating for Biodegradability Improvement 39
4. Kim, S., Lee, J., Kim, Y., Riu, D., Jung, S., Lee, Y.: Synthesis of Si, Mg substituted
hydroxyapatites and their sintering behaviors. Biomaterials 24, 1389–1398 (2003)
5. Dai, K.: Rational utilization of the stress shielding effect of implants. In: Poitout, D.G. (ed.)
Biomechanics and Biomaterials in Orthopedics, 1st edn. Springer, New york (2004)
6. Hulskes, R.: Stress shielding and bone resorption in THA : clinical versus computer-simula-
tion studies. Acta Orthop. Belg. 59(1), 118–129 (1993)
7. Sealy, M.P., Guo, Y.B.: Surface integrity and process mechanics of laser shock peening of
novel biodegradable magnesium-calcium (Mg-Ca) alloy. J. Mech. Behav. Biomed. Mater. 3,
488–496 (2010)
8. Li, Z., Gu, X., Lou, S., Zheng, Y.: The development of binary Mg-Ca alloys for use as biode-
gradable materials within bone. Biomaterials 29, 1329–1344 (2008)
9. Song, G.L., Atrens, A.: Corrosion mechanisms of magnesium alloys. Adv. Eng. Mater. 1,
11–33 (1999)
10. Razavi, M., Fathi, M.H., Meratian, M.: Bio-corrosion behavior of magnesium-fluorapatite
nanocomposite for biomedical applications. Mater. Lett. 64, 2487–2490 (2010)
11. Yang, J., Cui, F., Lee, I.: Plasma surface modification of magnesium alloy for biomedical
application. Surf. Coat. Technol. 205, S182–S187 (2010)
12. Fathi, M.H., Hanifi, A.: Evaluation and characterization of nanostructure hydroxyapatite
powder prepared by simple sol–gel method. Mater. Lett. 61, 3978–3983 (2007)
13. Bose, S., Dasgupta, S., Tarafder, S., Bandyopadhyay, A.: Microwave-processed nanocrys-
talline hydroxyapatite: simultaneous enhancement of mechanical and biological properties.
Acta Biomater. 6, 3782–3790 (2010)
14. Mohammadi Zahrani, E., Fathi, M.H., Alfantazi, A.M.: Sol-gel derived nanocrystalline fluor-
idated hydroxyapatite powders and nanostructured coatings for tissue engineering applica-
tions. Metall. Mater. Trans. A. 42, 3291–3309 (2010)
15. Ratner, B., Haffman, A. S., Schoen, F. J., Lemons, J. E.: Biomaterials Science: An
Introduction to Materials in Medicine, 2nd edn. Elsevier Academic Press, New york (1996)
16. Fathi, M.H., Hanifi, A., Mortazavi, V.: Preparation and bioactivity evaluation of bone-like
hydroxyapatite nanopowder. J. Mater. Process. Technol. 202, 536–542 (2008)
17. Bohner, M., Lemaitre, J.: Can bioactivity be tested in vitro with SBF solution? Biomaterials
30, 2175–2179 (2009)
18. Kokubo, T., Takadama, H.: How useful is SBF in predicting in vivo bone bioactivity?
Biomaterials 27, 2907–2915 (2006)
19. Fathi, M.H., Doostmohammadi, A.: Preparation and characterization of sol–gel bioactive glass
coating for improvement of biocompatibility of human body implant. Mater. Sci. Eng. A. 474,
128–133 (2008)
20. Stephen Tait, W.: An introduction to electrochemical corrosion testing for practicing engi-
neers and scientists. PairDocs Publications, Racine (1994)
21. Panda, R., Hsieh, M., Chung, R., Chin, T.S.: FTIR, XRD, SEM and solid state NMR investi-
gations of carbonate-containing hydroxyapatite nano-particles synthesized by hydroxide-gel
technique. J. Phys. Chem. Solids 64, 193–199 (2003)
22. Tadic, D., Epple, M.: Mechanically stable implants of synthetic bone mineral by cold iso-
static pressing. Biomaterials 24, 4565–4571 (2003)
23. Wei, M., Evans, J.H., Bostrom, T., Grøndahl, L.: Synthesis and characterization of hydroxyapa-
tite, fluoride-substituted hydroxyapatite and fluorapatite. J. Mater. Sci. Mater. Med. 14, 311–
320 (2003)
24. Medved, J., Mrvar, P., Vončina, M.: Oxidation resistance of AM60, AM50, AE42 and AZ91
magnesium alloys, Magnesium alloys-corrosion and surface treatments, Czerwinski, F. (ed),
InTech. http://www.intechopen.com/books/magnesium-alloys-corrosion-and-surface-treat-
ments/oxidation-resistance-of-am60-am50-ae42-and-az91-magnesium-alloys (2011)
25. Aung, N.N., Zhou, W.: Effect of grain size and twins on corrosion behaviour of AZ31B mag-
nesium alloy. Corros. Sci. 52, 589–594 (2010)
26. Song, G.: Control of biodegradation of biocompatible magnesium alloys. Corros. Sci. 49,
1696–1701 (2007)
40 R. Rojaee et al.
27. Witte, F., Hort, N., et al.: Degradable biomaterials based on magnesium corrosion. Curr.
Opin. Solid State Mater. Sci. 12, 63–72 (2008)
28. Kokubo, T.: Design of bioactive bone substitutes based on biomineralization process. Mater.
Sci. Eng C. 25, 97–104 (2005)
29. Zhang, S., Zhang, X., et al.: Research on an Mg-Zn alloy as a degradable biomaterial. Acta
Biomater. 6, 626–640 (2010)
The Effects of Cracks Emanating
from Micro-Void and Bone Inclusion
in Cemented Total Hip Replacement
M. M. Bouziane (*)
Department of Mechanical Engineering, University Mustapha Stambouli of Masacra,
Masacra, Algeria
e-mail: agk_bouziane@yahoo.fr
M. M. Bouziane · B. A. Bachir. Bouiadjra · B. Serier · S. Benbarek
LMPM, Department of Mechanical Engineering, University of Sidi Bel Abbes,
BP 89, City Ben M’hidi, Sidi Bel Abbes 22000, Algeria
B. Serier
e-mail: boualems@yahoo.fr
B. A. Bachir. Bouiadjra
Mechanical Engineering Department, College of Engineering, King Saud University,
Riyadh, Saudi Arabia
e-mail: bachirbou@yahoo.fr
N. Benseddiq
Laboratoire de Mécanique de Lille, IUT-A, 2 rue de la recherche,
59653 Villeneuve d’Ascq Cedex, France
e-mail: noureddine.benseddiq@univ-lille1.fr
E. M. H. Tabeti
Department of Orthopedic Surgery, University of Sidi Bel Abbes, City Ben M’hidi,
Sidi Bel Abbes 22000, Algeria
e-mail: chu-sba@sante.dz
Nomenclature
1 Introduction
Cemented total hip arthroplasty (THA) is the most popular method of osteoarthritis
treatment and other disorders affecting the hip joint [1]. Bone cement is based on
the brittle polymer polymethylmethacrylate (PMMA) and as a result, it is particu-
larly susceptible to stress concentrations [2]. The long-term stability of cemented
THA critically depends on the lasting integrity of the bond between bone, cement
and the stem. Polymethylmethacrylate (PMMA) bone cement has been widely
used for prosthetic fixation in orthopaedic surgery. PMMA gained popularity in the
early 1960 as a material for securing orthopaedic implants to bone [3]. Although
PMMA has been used successfully over the past three decades, there is a concern
that long-term looseing of the prosthesis may be related to mechanical failure of
the underlying cement mantle [4, 5]. Among problems encountered by experts
of cemented arthoplasty is the presence of defects in cement. These defects can
locally present a region of stresses concentrations producing a possible fracture of
cement and consequently the failure of the junction bone/implants. In general there
are three kinds of defects: porosities, inclusions and cracks [6]. In addition, there
are other discontinuities such as bone debris, blood entrapment, and interface irreg-
ularities, all of which could act as crack initiation sites [7].
One factor that could affect the fracture toughness is porosity, although some
investigators have suggested otherwise [8]. Since pores have been identified
in vitro as stress-risers and crack-initiators [9], higher degrees of porosity may
The Effects of Cracks Emanating 43
In this study, the finite element method is the tool employed to carry out the anal-
yses. The main reason to use this method is to predict the mode I stress inten-
sity factor (KI), the mode II stress intensity factor (KII) and the effective stress
intensity factor (Keff) of a cracks emanating from a micro-cavity and a crack ema-
nating from a bone inclusion. In this analysis the diameter of the micro-cavity
was assumed 200 μm, and the bone inclusion having the following dimensions
44 M. M. Bouziane et al.
Fig. 2 Crack emanating
from micro-void within the
bone cement
change its orientations from 0 to 360° by a step angle of 15° (Figs. 2 and 3). The
whole model is under a distributed compression loading of magnitude F = 30 N
[24]. The finite element code Abaqus 6.5.1 [25] was used to analyse the effect of a
crack in the cement. The mesh parameters are also introduced in order to control
element size and number in the various areas.
The stress intensity factor (SIF) at the crack front was extracted using the vir-
tual crack closure technique (VCCT). The virtual crack closure technique (VCCT)
criterion uses the principles of linear elastic fracture mechanics (LEFM) and,
therefore, is appropriate for problems in which brittle crack propagation occurs
along predefined surfaces. VCCT is based on the assumption that the strain energy
released when a crack is extended by a certain amount is the same as the energy
required to close the crack by the same amount [25] (Fig. 4).
Crack
(b)
(c)
(a)
Fig. 3 Mesh model (a) 2D FEM model, (b) crack emanating from micro-void, (c) crack tip
46 M. M. Bouziane et al.
Crack
(b)
(c)
(a)
Fig. 4 Mesh model (a) 2D FEM model, (b) crack emanating from bone inclusion, (c) crack tip
The idea presented by Rybicki and Kanninen is based on the calculation of the
energy release rate [26], using the Irwin assumption that the energy released in the
process of crack expansion is equal to work required to close the crack to its origi-
nal state as the crack extends by a small amount Δa. Irwin computed this work as:
∆a
W = u (r) .σ . (∆a − r) dr (1)
0
where u is the relative displacement, σ the stress, r the distance from the crack tip,
and Δa the change in virtual crack length.
Therefore, the energy release rate is:
∆a
W 1
G = lim = lim u (r) .σ (r − ∆a).dr (2)
∆a→0 ∆a ∆a→0 2.∆a
0
where Δa is the distance from the crack tip in the third direction and h is the ele-
ment size in the third direction .
To apply Eq. (5) to FE models comprising Eight-node brick elements, the inte-
grals in Eq. (5) are replaced with the sum:
2
1
G= Fki .uki (5)
2h∆a
k=1
where index i controls direction and index k controls the node b number. The
stress intensity factor is analyzed along the thickness of the specimen character-
ized by positions. The stress intensity factors KI, KII and KIII are respectively
the mode I, II and III crack propagation, usually used in linear elastic fracture
mechanics to characterize the local crack-tip/crack-line stress and displacement
fields. They are related to the energy release rate (the J-integral) through
1 T −1 (6)
G= K .B .K
8Π
where K = [KI, KII, KIII]T are the stress intensity factors and B is called the pre-
logarithmic energy factor matrix. For homogeneous, isotropic materials B is diag-
onal, and the above equation simplifies to
1 2 1 2
G= KI + KII2 + K (7)
Ē 2G III
where,
E = Ē for plane stress
and
E for plane strain, axi-symmetry, and three dimensions.
Ē = (1−θ )
Finite element analyses of cracked reconstruction were performed and stress
intensity factors KI and KII were calculated for each crack
√
σij π a
Kβ = β ; β = 1,2 (8)
fij (β)
where: Kβ is the stress intensity factor (SIF) for mode I and II with β = I, II; a is
the length of the crack; σij are the stresses distribution near the crack tip and fij is a
dimensionless quantity that depends on the load and geometry [27].
48 M. M. Bouziane et al.
The equation for the effective stress intensity factor for mixed mode problems,
suggested by Tanaka, is based on the dislocation model for fatigue crack propaga-
tion proposed by Weertman [28]:
3 Results
There is a relationship between the stability of the crack and its position, and for
crack propagation, the crack growth depending on the distance separating the
crack and the loading conditions [29]. One calculates the stress intensity factors
of the crack emanating from the bone inclusion and elliptical crack in the cement
mantle for different positions proximal, medial and distal of the cemented femo-
ral stem.
Before the calculation of stress intensity factors along the crack tip, it is neces-
sary to analyse the stress distribution around the bone inclusion without crack, in
order to predict the location of crack initiation. It was considered useful, to deter-
mine the distribution of the von Mises stress about normal and shear stresses in the
cement around the bone inclusion. In this study, a bi-dimensional finite elements
method was used to analyse the fracture behaviour of the cement, by computing
the stress intensity factors at the crack tip.
Finite element analyses of the prosthesis are carried out using ABAQUS. The dis-
tribution of the von Mises stresses in the cement mantle around the micro-void
and crack tip for different positions (proximal, medial and distal zones) are illus-
trated in Fig. 5. The stress distribution is not uniform around the micro-defect.
For different positions, the maximum stresses are located around the crack front.
These high stresses may lead to crack propagation in the cement. These cracks
may lead to loosening of the prosthesis. Comparing the stress distribution, the
highest stress in the cement mantle around the tip crack is located in the distal
part, the stresses levels are very significant, and they can reach 24 MPa. The stem
tip action on the distal part of the cement increased the stresses. For the medial
part, the stresses level are important, the maximum stress value is 16 MPa. Finally,
in the proximal part, the stresses around the tip crack are lower than the two other
regions. Thus, the presence of the micro-defect in this position presents the least
dangerous case.
The Effects of Cracks Emanating 49
micro-void
Crack
(a)
(b)
(c)
Fig. 5 Von Mises stresses (MPa) in the cement mantle around the micro-defect (crack emanat-
ing from micro-void) different positions: a proximal part, b medial part, c distal part
Figures 6 and 7 present the variation of KI and KII according to a crack emanat-
ing from a micro-cavity in the cement mantle of a different part of THA (proxi-
mal, medial and distal). Based on the numerical analysis it can be found that an
orientation of the crack from 0° to 90° and -90° to -180° gives positive values of
KI. The opening mode reaches the maximum value when the crack is tilted to 45°
and -135° for the three zones that indicates the dangerous orientations angles. The
obtained results show that the stress intensity factor is null for the remaining crack
inclination angles and this are independently of the posture of the patient. This
behaviour can be explained by the nature of the stress field induced in cement of
the micro-cavity.
Figure 8 presents the computed effective stress intensity factors (Keff) of a
crack emanating from a micro-void for different positions. The effective stress
intensity factors were found decreasing and increasing around the micro-void for
all cases. It is clearly shown that the Keff for a crack in the distal part are higher
50 M. M. Bouziane et al.
0,20
Proximal part
KI (MPa m )
Medial part
1/2
Distal part
0,15
0,10
0,05
angle ( α )
0,00
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180
Fig. 6 Stress intensity factor KI variation according to crack orientation within the cement mantle
KII (MPa m )
Proximal part
1/2
Fig. 7 Stress intensity factor KII variation according to crack orientation within the cement
mantle
comparing the others, the maximum value is about 0.5 MPam1/2 for a crack angle
of α = -75° (see Table 3). Concerning the medial zone, the maximum Keff value
can reach 0.3 MPam1/2 for a crack orientation of α = -90°. For the proximal part,
it is found that the effective stress intensity factor is relatively weak, the Keff val-
ues lie between 0 and 0.2 MPam1/2.
The Effects of Cracks Emanating 51
Proximal part
0.4
0.3
0.2
0.1
angle (α )
0.0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180
Fig. 8 Effective stress intensity factor Keff variation according to crack orientation within the
cement mantle
Table 3 Comparison of maximum SIF (KI, KII, Keff) in the different zones of crack emanating
from the micro-void
Maximum mode I Maximum mode II Maximum effective
SIF (MPa m1/2) SIF (MPa m1/2) SIF (MPa m1/2)
Proximal position 0,08035 0,10325 0,17365
Medial position 0,12007 0,17133 0,28814
Distal position 0,17766 0,27733 0,51374
Figure 9 presents the distribution of the von Mises stresses in the cement mantle
around the bone inclusion and crack tip for different positions (proximal, medial
and distal zones). For distal part, the level of stresses are very important, they can
reach 25 MPa. This value permits us to conclude that the risk of crack initiation by
bone inclusion is higher. For the medial zone, the maximum stress value is 18 MPa
in the cement. Proximally, the stresses around micro-defect are lower than the two
other regions; the maximum stress value is about 11 MPa.
52 M. M. Bouziane et al.
Bone
inclusion
Crack
(a)
(b)
(c)
Fig. 9 Von Mises stresses (MPa) in the cement mantle around the micro-defect (crack emanat-
ing from bone inclusion) different positions: a proximal part, b medial part, c distal part
Figures 10 and 11 reveal the evolution of the stress intensity factor in mode I (KI)
and mode II (KII), according to the crack inclination angle starting from a bone
inclusion, in the three positions (proximal, medial and distal), the same behaviour of
both KI and KII are seen. The opening mode KI is positive between the crack inclina-
tion angle 0° and 90°, and the crack closed for the inclination range 0° and -90°. The
shearing mode (KII) takes negative values for the crack inclination angle reported
between -45° and 45° and positive for the remaining orientations. The higher values
of shearing mode are noted when the cracks are inclined by 90° and -90°.
The numerical results are illustrated in Fig. 12, its show the variation of the
effective stress intensity factors according to a selected crack inclinations angle α
emanating by bone inclusion, in the three parts of the external side of THA (proxi-
mal, medial and distal parts). It is clearly shown that the Keff for a crack in the dis-
tal part are higher comparing the others, the maximum value is about 0.47 MPam1/2
for crack angle α = 45° (see the Table 4). Concerning the medial zone, the maxi-
mum Keff value can reach 0.4 MPam1/2 for crack orientation α = 45°.
The Effects of Cracks Emanating 53
0,4
Partie proximale
KI (MPa m )
1/2
0,3
Partie médiale
Partie distale
0,2
0,1
angle ( α )
0,0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180
-0,1
-0,2
-0,3
-0,4
Fig. 10 Stress intensity factor KI variation according to crack orientation within the cement mantle
Proximal part
Medial part
KII (MPa m )
0,5
1/2
Distal part
0,4
0,3
0,2
0,1
angle ( α )
0,0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180
-0,1
-0,2
-0,3
-0,4
-0,5
-0,6
Fig. 11 Stress intensity factor KII variation according to crack orientation within the cement
mantle
For the proximal part, it is found that the effective stress intensity factor is rela-
tively weak, the Keff values lie between 0 and 0.1 MPam1/2.
54 M. M. Bouziane et al.
Proximal part
Medial part
Keff (MPa m )
1/2
0,5 Distal part
0,4
0,3
0,2
0,1
angle ( α )
0,0
Fig. 12 Effective stress intensity factor Keff variation according to crack orientation within the
cement mantle
Table 4 Comparison of maximum SIF (KI, KII, Keff) in the different zones of crack emanating
from the bone inclusion
Maximum mode I Maximum mode II Maximum effective
SIF (MPa m1/2) SIF (MPa m1/2) SIF (MPa m1/2)
Proximal position 0.06004 0.17365 0.12457
Medial position 0.24035 0.29654 0.38578
Distal position 0.28760 0.48452 0.47054
4 Discussions
Micro-cracking may lead to PMMA, which can induce local inflammation and
osteolysis [17, 30]. The relative position, the geometry and size of any cracks
have significant effect on the propagation of the crack. Similar results have
been described by other authors. McCormack found that the cracks which initi-
ate in the cement mantle early on in the loading history occur at locations of
high stress location [19]. Byeongsoo Kim found that the SIF values decreasing
from 0.2 to 0.37 MPa m1/2, this suggests that the propensity for crack propaga-
tion will decrease as the crack grows in the cement mantle [31]. Michael D. Ries
found that the critical stress intensity factor KIC lies between 0.96 MPa m1/2 to
1.76 MPa m1/2 [32].
Tables 3 and 4 show respectively the comparison between the variation of stress
intensity factors (KI, KII and Keff) for different positions (proximal, medial and
The Effects of Cracks Emanating 55
distal) of a cracks emanating from micro-void and cracks emanating from bone
inclusion, in the cement mantle of the THA. one can see that, the cracks existing
in the distal part lead to the highest values of SIF.
In general, cracks can inhibit or promote propagation of adjacent cracks
depending on their relative position, size and the degree of interaction between the
induced crack tip concentrations. The results of this analysis enable us to provide
solutions to stop the crack propagation in the bone cement, as using a grid in the
cement mantle and reinforce cement in the crack path, one use nanotechnology to
create area with higher mechanical properties.
5 Conclusion
The primary objective of this study was accomplished; to evaluate stresses con-
centration around the micro-defect (micro-void and bone inclusion) and the stress
intensity factors in the three parts of THA of a crack emanating from a micro-
defect in the cement mantle; the following conclusions can be made:
- The micro-defect located in the distal part of the cement presents the most
important risk of the fracture of the cement; the effect of the implant tip for this
region of the cement increases the stresses. The micro-defect situated in the medial
part presents also the risk of the rupture of the cement mantle. The micro-defect
situated in the proximal part is subjected to weak stresses; therefore, it does not
present an important risk of the rupture.
- Stress intensity values in mode I, mode II and Keff depends on the orientation
of crack emanating from a micro-defect. The distal part is the most dangerous part
that gives high values of SIF in opening and shearing modes, in the presence of
a crack emanation of a micro-defect, in this zone for a vertical crack the damage
risk by brutal rupture is significant. The rupture risk by the shearing mode is more
significant for a micro-defect located in the distal part of orthopaedic cement.
Acknowledgments The Authors extend their appreciation to the Deanship of Scientific
Research at King Saud University for funding the work through the research group No. RGP-
VPP-035.
Refernces
1. Huiskes, R.: Some fundamental aspects of human joint replacement. Analyses of stresses
and heat conduction in bone- prothesis structures. Acta. Orthop Scand. 185 (Supplément n),
(1980)
2. Hoey, D., Taylor, D.: Fatigue in porous PMMA: the effect of stress concentrations. Int. J.
Fatigue 30(6), 989–995 (2008)
3. Charnley, J.: Total hip replacement by low-friction arthroplasty. Clin. Orthop 72(7), 21
(1970)
4. Saha, S., Pal, S.: Mechanical properties of bone cement: a review. J. Biomed. Mater. Res
18(435), 62 (1984)
56 M. M. Bouziane et al.
5. Lewis, G.: Properties of acrylic bone cement: state of the art review. J. Biomed. Mater. Res
38(155), 82 (1997)
6. Bhambri, S.K., Gilbertson, L.N.: Micro mechanisms of fatigue crack initiation and propaga-
tion in bone cements. J. Biomed. Mater. Res 29, 233–237 (1995)
7. Murphy, B.P., Prendergast, P.J.: The Relationship Between Stress, Porosity, and Nonlinear
Damage Accumulation in Acrylic Bone Cement. J. Biomed. Mater. Res 59, 646–654 (2001).
Department of Mechanical Engineering, Trinity College, Dublin 2, Ireland
8. Vila, M.M., Ginebra, M.P., Gil, F.J., Planell, J.A.: Effect of porosity and environment on the
mechanical behavior of acrylic bone cement modified with acrylonitrile-butadiene-styrene
particles: I. Fracture toughness. J. Biomed. Mater. Res 48(2), 121–127 (1999)
9. Topoleski, L.D., Ducheyne, P., Cuckler, J.M.: Microstructural pathway of fracture in poly
(methyl methacrylate) bone cement. Biomater 14(15), 1165–1172 (1993)
10. Bhambri, S.K., Gilbertson, L.N.: Micromechanisms of fatigue crack initiation and propaga-
tion in bone cements. J. Biomed. Mater. Res 29, 233–237 (1995)
11. McCormack, B.A.O., Prendergast, P.J., Gallagher, D.J.: An experimental study of damage
accumulation in cemented hip prostheses. Clin. Biomech 11(4), 214–219 (1996)
12. Murphy, B.P., Prendergast, P.J.: The relationship between stress, porosity, and nonlinear dam-
age accumulation in acrylic bone cement. J. Biomed. Mater. Res 59(4), 646–654 (2002)
13. Dunne, N.J., Orr, J.F.: Influence of mixing techniques on the physical properties of bone
cement. Biomater 22(13), 1819–1826 (2001)
14. Graham, J., Pruitt, L.A., Ries, M.D., Gundiah, N.: Fracture and fatigue properties of acrylic
bone cement: the effects of mixing method, sterilization treatment, and molecular weight. J.
Arthroplasty 15(8), 1028–1035 (2000)
15. Harper, E.P., Braden, M., Bonfield, W., et al.: Influence of sterilization upon a range of prop-
erties of experimental bone cements. J. Mater. Sci. Mater. Med 8, 849 (1997)
16. Lewis, G., Mladsi, S.: Effect of sterilization method on properties of Palacos Racrylic bone
cement. Biomater 19(1–3), 117–124 (1998)
17. Hughes, K.F., Ries, M.D., Pruitt, L.A.: Structural degradation of acrylic bone cements due to
in vivo and simulated aging. J. Biomed. Mater. Res 65A(2), 126–135 (2003)
18. Hertzberg, R.W.: Deformation and fracture mechanics of engineering materials, 2nd edn.
Wiley, New York (1983). [chapters 7–8]
19. McCormack, B.A.O., Prendergast, P.J.: Microdamage accumulation in the cement layer of
hip replacements under flexural loading. J. Biomech. 32(5), 467–475 (1999)
20. Bachir Bouiadjra, B., Belarbi, A., Benbarek, S.,et al: FE analysis of the behaviour of micro-
cracks in the cement mantle of reconstructed acetabulum in the total hip prosthesis.Comput.
Mater. Sci. 40(4), 485–491 (2007)
21. Harrigan, T.E., Harris, W.H.: Mechanical consequences of cement-metal interface disruption
(debonding) at thedistal tip of femoral total hip prosthesis. Trans. Orthop. Res. Soc 16, 503
(1991)
22. Estok, D.M., Harrigan, T.E., Harris, W.H.: Finite element analysis of cement strains at the tip
of an idealized cemented femoral component. Trans. Orthop. Res. Soc 16, 504 (1991)
23. Lee, I.Y., Skinner, H.B., Keyak, J.H.: Effects of variation of prosthesis size on cement stress
at the tip of a femoral implant. J. Biomed. Mater. Res 28, 1055 (1994)
24. Bergmann, G., Graichen, F., Rohlmann, A.: Hip Joint Loading Measured During Walking
and Running, Measured in Two Patients. J. Biomech 26, 969–990 (1993)
25. ABAQUS Ver 6-5, User guide, Cornell University, (2004)
26. Rybicki, E.F., Kanninen, M.F.: A finite element calculation of stress intensity factors by a
modified crack closure integral. Eng. Fract. Mech 9, 931–938 (1977)
27. Irwing, G.R.: Analysis of stress and strain near the end of a crack traversing a plate. J. Appl.
Mech. ASME 24, 361–364 (1957)
28. Weertman, J.: Rate of growth of fatigue cracks calculated from the theory of infinitesimal
dislocations distributed on a plane. Int. J. Fract. Mech 2, 460–467 (1966)
29. Eberhardt, E., Stead, D., Stimpson, B.,et al. The effect of neighbouring cracks on elliptical crack
initiation and propagation in uniaxial and triaxial stress field. J. Eng.Fract. Mech. 59: 103–115
The Effects of Cracks Emanating 57
30. Horowitz, SM., Doty, SB., Lane, JM.,et al.: Studies of the mechanism by which the mechani-
cal failure of polymethylmethacrylate leads to bone resorption. J Bone Jt Surg. 803–813
(1993)
31. Byeongsoo, K., Byungyoung, M., Kenneth, AM.,et al.: Simulated crack propagation in
cemented total hip replacements. Mater. Sci. Eng. 0921-5093 (2007)
32. Michael, DR., Ernest, Y., Laila, A., et al.: In vivo behavior of acrylic bone cement in total hip
arthroplasty. Biomater. 256–261 (2006)
Safety Measures for Avoiding or Mitigating
the Occupant Exposure in Collisions with
Large Animals
Abstract Road accidents involving large animals are responsible for a c onsiderable
number of fatalities; these are e.g. elk in Northern Europe or camels in Middle
East, Horn of Africa and Maghreb. In order to address this topic, investigations
concerning active and passive road safety have been launched at the King Saud
University and at Graz University of Technology. Data of real accidents or acci-
dent statistics are not available in the required quality to gather typical accident sce-
narios. Therefore generic accidents types were derived and critical load cases were
determined by means of multi body and finite element simulations. It was found that
the vehicle acceleration due to the impact is not a safety issue, but the finite ele-
ment simulations showed that massive local intrusions into the passenger compart-
ment are the main cause for severe occupant injuries. Therefore, concepts for the
prevention of critical intrusions were examined. In a first step the space required
for the occupant movement due to the rather mediocre acceleration levels in car-to-
large animal accidents was determined by multi-body simulation. In a second step
different approaches preventing intrusions into the occupant-movement-space were
analyzed by finite element simulation. In addition, the possible effects and require-
ments of active systems have been studied to demonstrate the potential of combined
measures.
1 Literature Study
In the literature there is only few data about accidents with large animals, espe-
cially detailed analysis regarding this accident type is missing. In the early 1990s
Bruinderink and Hazebroek investigated accidents with hoofed animals in Europe,
whereas finally the data of eight countries (Austria, Denmark, Finland, Germany,
Ireland, Norway, Sweden and the Netherlands) could be evaluated. The expan-
sion from the data on hand showed for Europe (excluding Russia) a value of over
500,000 accidents with 300 killed and approx. 30,000 injured persons and a dam-
age sum of over a billion Euros [1]. More current investigations show, for exam-
ple, between the years 2000 and 2005 for Germany alone 210,000−235,000 deer
accidents every year that are registered by the insurance companies. The annual
regulatory costs added up to an average of 370−464 billion Euros. Although only
damage to property is caused by these accidents, accidents with large animals are
a danger to vehicle occupants. German statistics for accidents with animals for
this period show between 18 and 28 fatally injured persons and between 580 and
750 severely injured and 2,100−2,600 light casualties every year [2]. For the years
1979−1980 the official Swedish statistics registered 900 passenger car/deer acci-
dents with bodily injuries [3].
In the US approx. 200 persons are fatally injured in accidents involving ani-
mals (FARS 2002−2007). It is estimated that in the years 2001 and 2002 approx.
27,000 persons had to be treated for non-fatal injuries because of accidents involv-
ing animals (Centers for Disease Control, 2004) [4]. The Insurance Institute for
Highway Safety (IIHS) reports that in the US approx. one and a half million acci-
dents with red deer occur, causing a damage sum of at least 1.1 billion US Dollars
and about 150 killed persons and the numbers are rising [5].
Safety Measures for Avoiding or Mitigating the Occupant 61
In the near East there are only a few statistical data, most of the accidents are
with camels. Al-Ghamdi and AlGadhi reported in 2004 of 600 camel-vehicle colli-
sions per year in Saudi Arabia with a death rate of 0.25 per accident, representing
six times as much as all other types of accidents [6]. Attewell and Glase reported
from Australia 94 fatal and 1,392 accidents (between 1990 and 1997) causing
people to be hospitalized. These facts are based on data gained by the Australian
National Transport Agency, whereas for 71 % of the fatal accidents large animals
such as cattle or horses were documented [7].
One factor is valid for all investigated countries. The main cause for vehicle-
animal accidents with fatally or severely injured persons are large animals. In the
US it is the accidents with red deer in Europe and Canada it is elk and red deer, in
Australia cattle and horses and in the Near East it is camels.
Most of the accidents are head-on collisions, i.e. the initial contact is with
the car’s front. Especially when large animals collide with a motor car only the
legs are caught by the vehicle front whereas the torso hits in most cases the wind
screen, the A-pillar and the roof causing the most significant deformations. These
intrusions lead to injuries of the vehicle occupants because the structure of the
vehicle is not designed for this kind of load. Dependant on the investigation meth-
ods also a considerable proportion of fatal accidents and accidents with severely
injured is the result, whereas the evasive manoeuvres to dodge the blow caused the
accident. It is as good as impossible to draw definite conclusions when there has
been no contact with an animal and there is no real evidence available and the size
of the animal plays a rather subordinate role. In addition there is a certain amount
of accidents where the initial contact with the animal took place on the road space
leading to a follow-up impact with the infrastructure (tree, pole, bank etc.) or to a
rollover. These cases are not taken into account for investigations but methods of
prevention (especially in regard to the infrastructure and the active vehicle safety)
can surely have a positive effect on this type of accident [8].
2 Method
The principal steps for the method, which was used for this analysis, is shown in
Fig. 1. The first step was to develop the relevant accident scenario, which was done
using statistics, literature and simulation as basis. For the analysis of the safety
measures, combinations of multibody and finite elements (FE) methods were used.
Therefore, suitable vehicle and animal models had to be developed, which are
suitable to represent the situation correctly, but also fast enough to perform vari-
ations at reasonable time. Through simulation and optimization measures were
developed to minimize the risk for the vehicle occupants. Both, passive safety
measures as well as active safety measures were examined. Finally it was shown
that by a combination of suitable active and passive measures the risk for occupant
injuries can be minimized.
62 W. Sinz et al.
Fig. 1 Method used for the analysis to get typical accident scenarios and investigate safety measures
to avoid these accident types or reduce the injuries caused by accidents with large animals
3 Accident Type
On the basis of literature research and statistics an attempt was made to extract
typical accident scenarios involving animals. However, most data are not detailed
enough to develop specific accident scenarios. Therefore, an analysis was made by
simulations to define worst case scenarios, where one can assume a risk to vehicle
occupants.
Database research and analysis of real world accidents, with focus on fatal and
severe accidents, showed, that frontal collisions are the main type of accident,
which means that the initial contact between vehicle and animal occurs with the
vehicle front. But there is not enough information about the configuration, e.g. the
velocity of the animal or the crossing angle of the animal.
Dependent on the investigation methods also a considerable proportion of fatal
accidents and accidents with severely injured are the result of evasive maneuvers
to avoid the accident. It is as good as impossible to draw definite conclusions
when there has been no contact with an animal and there is no real evidence avail-
able and the size of the animal plays a rather subordinate role. In addition, there is
a certain amount of accidents where the initial contact with the animal took place
on the road space leading to a follow-up impact with the infrastructure (tree, pole,
bank etc.) or to a rollover. These cases are not taken into account for these inves-
tigations but methods of prevention (especially in regard to the infrastructure and
the active vehicle safety) can surely have a positive effect on this type of accident.
The basic procedure is that from a database the corresponding type of acci-
dent is identified by filtering and by evaluating the relevant parameters conclu-
sions can be made to the accident configuration. The problem is that for in-depth
analysis the official statistics do not provide enough information. The necessary
information for in-depth analysis can only come from appropriately evaluated
Safety Measures for Avoiding or Mitigating the Occupant 63
real accidents. To close the gap between general data from official statistics and
the necessary information for detailed analysis at the Vehicle Safety Institute of
Graz University of Technology a separate database was established, the ZEDATU
(Central Database for in-depth analysis of traffic accidents) [8, 9].
The principal approach was that we used the national statistics to select animal
accidents, than we selected cases from the ZEDATU, which is possible because
both database are linked. The real accidents were reconstructed and analyzed to
get relevant parameters. A generic accident was derived and through parameter
variation the influence of different parameter on the risk for the occupants could
be determined (Fig. 2).
Therefore, studies were made with the multi-body model of the accident recon-
struction program PC-Crash. Different types of vehicles and various types of ani-
mals were modeled and different accident configurations were analyzed, where the
parameter type of vehicle, type of animal, vehicle speed, animal velocity, crossing
angle of the animal were varied (Fig. 3).
Based on these variations two typical configurations which are of relevance for
the safety of the vehicle occupants could be detected. In this case, however, the
effect on the driving dynamics was for this study of minor importance, that is e.g.
when the vehicle was influenced by the impact in such a way that a critical driving
condition could arise eventually, but the real impact is not critical.
Based on these variations two typical configurations, where the impact is with
the vehicle’s front, were detected.
In principle two kinds of accidents can be distinguished, if the focus is on fron-
tal impact:
• Type A: Collisions with animals affecting only the front area of the vehicle and
no contact with passenger cell (Fig. 4).
• Type B: Collisions with animals, where there is contact to the passenger cell
due to the body geometry (Fig. 5).
Concerning traffic safety there are principally three components of interest, the
human, the vehicle and the environment. In Fig. 6 the phases of an accident are
shown and examples for influencing factors are listed. The interaction between
these three components cannot be solved only technically, but also other factors
such as policy, psychology, design, etc. play a role.
In principle, one must distinguish between active safety measures with regard
to measures concerning the vehicle, the driver and the infrastructure. These meas-
ures have already been treated by the authors in a publication [10]. Therefore, in
this study only vehicle-related measures are discussed.
For accident configuration one, frontal collision without contact with the pas-
senger cell (Fig. 4), the load is not the imminent danger for the occupants. One can
recognize that when smaller animals are involved the greatest danger comes up as
due to the collision the vehicle dynamic is changed and critical driving situations
bearing the danger of a secondary collision. In addition there is the danger of an
animal running around on the road area which can cause an evasive maneuver by
the driver. This can lead to a dangerous driving situation even without a collision
with the animal. For this situation there are different avoiding strategies, mainly
active measures. As partly seen already in modern vehicles e.g. active light, break-
ing assistant, object detection, antilock braking system (ABS), electronic stability
program (ESP) etc. In order to avoid or to cope with such situations concerning the
driver, especially trainings and awareness rising can be mentioned here. By focused
training of crucial situations like vehicle engineering courses or awareness raising
and practicing appropriate measures, handling the vehicle and reacting in a danger
Safety Measures for Avoiding or Mitigating the Occupant 65
situation can be improved. In addition sensitization of the vehicle driver for specific
situations can be enhanced which should lead to a better estimation of situations
and more attention and passive driving. In connection with measures for infrastruc-
ture, safety measures that stop animals from crossing the road are relevant.
In the case of larger animals there are not only the dangers described above but
there is the risk that the animal or parts of it touch the windscreen, the A-pillar
or the front part of the roof and thus causing an intrusion into the passenger cell
(Fig. 5) causing fatal injuries, or serious injuries in the area of the head, spine and
thorax of the passengers. This can be seen when parts of the animal’s body lie
above the vehicle front and the collision causes the vehicle to rotate. When large
animals collide with a motor car only the legs are caught by the vehicle front
whereas the torso hits in most cases the wind screen, the A-pillar and the roof.
Most significant deformations are caused and these intrusions lead to injuries of
the vehicle occupants because the structure of the vehicle is not designed for this
kind of load, because there are no compulsory tests or load cases. Measures to
mitigate this kind of accident configuration are referred to passive safety, e.g. rein-
forcement of the A-pillar and other constructive measures to avoid or mitigate an
impact on the passenger cell. This situation does not only apply to micro cars or
small cars (compact cars) but all vehicles where the geometrical conditions con-
cerning the vehicle front end and the animal allows that the animal or parts of the
body hits the occupant compartment. But also active safety and combinations of
active and passive safety measures can be suitable to minimize the outcome of
such accident configurations.
In principle in vehicle safety one can differ between active safety, anything that
helps to avoid the accident, and passive safety, measures that reduce the conse-
quences of an accident (Fig. 7). The principle of integrated vehicle safety is based
on the networking of active and passive safety systems into a comprehensive over-
all safety strategy, which is selected according to the current environment situa-
tion. In the following active systems requirements are discussed for, although, as
already mentioned, received only on measures which affect the vehicle.
Intelligent vehicle systems can basically be distinguished between three catego-
ries [12]:
• Systems which provide an advisory/warning to the driver (collision warning
systems)
• Systems which take partial control of the vehicle (driver assistance, collision
avoidance)
• Systems which take full control of vehicle operation (vehicle automation).
The systems can work as autonomous systems, with all necessary instrumenta-
tion and intelligence on-board or as cooperative systems, in which assistance is
Safety Measures for Avoiding or Mitigating the Occupant 67
provided from other vehicles, the infrastructure or other sources (Car-to-Car com-
munication ‘C2C’, Car-to-Infrastructure ‘C2I’, Car-to-X communication ‘C2X’).
In addition to a variety of systems that are currently used in modern vehicles,
for accidents with large animals are especially anti-lock braking system, electronic
stability control, brake assist system, light source technology, night vision systems
and identification of objects of interest.
The aim of this study was not to develop an appropriate algorithm to detect
animal accidents, but to define the requirements for a sensor in order to avoid
accidents. Requirements and possibilities of the various active systems should be
integrated into a classification scheme. This classification scheme should point out
the possible benefits which could be achieved using active systems, or the change
regarding the avoidance of different accident types. The determination of the
effectiveness of driving assistance systems and the control switch for driver assis-
tance systems are the topics of interest, respectively how should the environment
appear in order to avoid accidents with animals.
For collisions with animals, object detection in combination with an automatic
brake application would be suitable. Therefore, accident scenarios with differ-
ent impact configuration and impact velocity were analyzed, using the accident
reconstruction software PC-Crash [13]. Because of the lack of statistical data
and real world cases one typical scenario (Fig. 8) was assumed to point out the
principal approach. Further investigations are necessary in regard to real accident
events in order to assess which sensors or which combination of sensors delivers
the most effective solution, especially in combination with constraints from the
environment.
For this accident scenario the velocity of the vehicle was 100 kph and the camel
was crossing with 10 kph. The emergency braking was initiated 0.2 s before the
collision, assuming a reaction time of 0.8 s, the reaction was 1.0 s before the col-
lision, at a distance of 28 m before the actual collision point. In this 1 s the camel,
68 W. Sinz et al.
Fig. 8 Accident scenario, 100 kph 90° impact vehicle against camel, camel crossing the road
with 10 kph from left
Fig. 9 Sequences of the accident, 100 kph 90° impact vehicle against camel, camel crossing the
road with 10 kph from left
which had a moving velocity of 10 kph, moved over a distance of approx. 3 m.
This means that the camel had already moved approx. 3 m on the road, when the
reaction was carried out of the vehicle driver (Fig. 9).
For example, if the vehicle had an active system, which could identify the
camel at a time (Fig. 10), when it enters the street and the reaction of an active
brake system is 0.2 s after object identification the accident could be avoided
(Fig. 11).
For defining sensor requirements a kinematic analysis was made, defining the
necessary sensor range for avoiding collisions with animals. The basic assump-
tions are that an animal is crossing the street with a defined velocity and angle.
The sensor is defined ideal, it means that the animal is detected without loss of
Safety Measures for Avoiding or Mitigating the Occupant 69
time and there also is no brake lag. The width of the vehicle and the length of the
vehicle are not considered. Within Fig. 12 the graphs for different vehicle veloci-
ties are prepared. For example if the driving velocity of the vehicle is 100 kph and
with the ideal sensor and brake as explained above, the necessary sensor field is
31 m if the animal is running in the same direction with 10 kph. If the animal is
moving with 10 kph in the opposite direction, oncoming to the vehicle, and the
vehicle should be stopped at the point of impact, the sensor range must be mini-
mum 46 m. The maximum lateral sensor requirement is when the animal crosses
the road at an angle of 90° (black bold line in Fig. 12), in the case of an animal
with 10 kph and a vehicle with 100 kph the sensor range must be at a minimum of
longitudinal app. 39 m and lateral 8 m.
If there is a brake lag time, e.g. 0.2 s, for a vehicle with a velocity of 100
kph and an animal walking with 10 kph the longitudinal sensor field must be
increased more than 5 m and the lateral sensor range approx. 0.6 m. Additional
longitudinal and lateral extension is required dependent on the time for object
recognition. It can be clearly seen that such an approach with sensors, object
detection and brake assistant is only partially meaningful, or that it is hardly
70 W. Sinz et al.
Fig. 12 Sensor requirement for avoiding accident with animal, animal with velocity of 10 kph at
different crossing angles, vehicle braking with 10 m/s²
feasible as pure active safety measure, since for example for 100 kph a lateral
stripe along the road with about 10 m is needed, and additionally this area should
be flat and free of other objects. Additionally the limits trough the coefficient of
friction must be observed. In the example above an optimum braking decelera-
tion with 10 m/s² was presumed, reduction of the coefficient of friction (e.g. wet
road, material on the road…) result in relevant higher values for the sensor field.
For example, if the vehicle’s deceleration is only 5 kph and an animal is crossing
the road at an angel of 90°, the sensor field needed to stop the vehicle is longi-
tudinal 77 m and lateral 15 m. One can easily see that the sensor needs to bring
the vehicle to rest are influenced by various boundary conditions. But in addition
the severity of the accident can be reduced by the use of sensors with object rec-
ognition and combined braking intervention, even if the accident itself cannot be
avoided by the system.
Some systems are already in use in various modern vehicles and more sys-
tems are being developed and tested which should have the potential to miti-
gate or avoid accidents with animals. The release of full braking and other
safety means depend on the quality of object detection and this subject was
not part of this investigation. Requirements of sensors and safety systems have
to be defined according to the intended purpose of covering all measures of
the accident events and a combination of various means would certainly be an
improvement.
More suitable models have to be developed and defined to calculate the require-
ments for vehicles. Defining the requirements for sensors need a lot more detailed
investigation.
Safety Measures for Avoiding or Mitigating the Occupant 71
The problems within these investigations are that there are only poor data for the
validation of the animals available; these data can only be obtained by separate
tests. Therefore, for the definition of the bodies as well as the stiffness and joint
definition of the animals merely estimated values were taken.
The mass ratio between animal and vehicle and the geometric situation are deci-
sive parameters for the change of vehicle velocity at the point of impact and therefore
it is a factor for the load on the passenger cell. But depending on geometrical condi-
tions it is possible that only a part of the animal hits the vehicle front and other parts
of the animal hit the passenger cell, especially at the windscreen, A-pillar and the
front part of the roof. Through massive deformations and direct contact of the passen-
gers with parts of the vehicle structure and the animal there is a high injury potential.
The method used for analysis and improvements regarding passive safety is
shown in Fig. 13. In a first step a simulation matrix with different configurations
was established to identify the worst case scenario through FE simulations.
The second step was the investigation of the occupant’s load, using a combina-
tion of FE-simulation and multibody simulation. The FE simulation was done for
the load case US-NCAP and the load case ‘sitting animal’ to calculate the vehicle
acceleration, which was used for a multibody simulation to investigate the loads
on the occupants.
Fig. 13 Method used for the analysis and improvements regarding passive safety
72 W. Sinz et al.
The third step was the investigation of the ‘concept cage’, which should reduce
the deformations into the occupant compartment. For this the ‘moving space’ of
the occupant within the US-NCAP load case was used to define a ‘safety area’,
where no intrusions should occur.
To recognize the vulnerability and to identify the mechanisms finite element
simulations were performed. Therefore for this analysis a vehicle was used, where
a suitable model and FE mesh (Fig. 14) were available. Due to the lack of informa-
tion regarding statistical data a simulation matrix was created, which was derived
from the PC-Crash simulation for the assessment of the accident type, with veloc-
ity of vehicle and animal and impact angle of the animal as variation parameter.
For the animal the FE models of a camel and of a moose were used. The camel
model has a mass of 450 kg (Fig. 15), left side), the moose model has a mass of
350 kg (Fig. 15), right side).
A simulation matrix was established with a set of 48 different simulation con-
figurations, with different configurations:
• Fully/partially covered
• Different obstacle (camel, moose)
• Angle of the obstacle (0°, 45°, 90°)
• Velocity of the obstacle (0 kph, 10 kph)
• Velocity of the car (50 kph, 100 kph).
In Fig. 16 one of the impact configurations can be seen. There were differences
of the mass of the animals, the moose model had a mass of 350 kg, the mass of
the camel was 450 kg, and also geometric differences, but the simulation results
were similar, there were slight deformations at the front bumper but relatively
Fig. 15 FE-model of camel (left side) and FE-model of the moose (right side)
Fig. 16 Example for a crash configuration, in this case a moose impact with 90° angle between
moose and vehicle and impact at the middle of the vehicle front (left side) and the corresponding
vehicle deformations
heavy deformations at the A-pillar and the roof. This can be justified that the body
or torso of the animal with a majority of the mass is higher than the front end.
Therefore at the initial impact mainly the animal’s legs are hit, the torso of the
animal is thereby caused to rotate and the vehicle continues to move toward torso,
with a further impact in the area of the A-pillar and the roof.
The kinematics is very similar for both, the camel and also the moose, but
for the load on the vehicle there are slight differences, for example are the legs
of the moose are heavier and tighter than the camel’s legs. Therefore within the
first impact against the front of the vehicle the vehicle is more decelerated in com-
parison to the impact against the camel. For example see the velocity time graph
during a 90° impact against a moose with 100 kph and a camel with 100 kph.
Figure 17 shows the vehicle velocity in longitudinal direction of the vehicle, the
red graph for the moose impact and the blue for the camel impact; the impact con-
ditions were the same for both animals. At first impact with the front the vehicle
is more decelerated within the moose accident compared to the camel, but at the
74 W. Sinz et al.
Fig. 17 Velocity of the car in longitudinal direction, moos vs. camel, fully covered, vehicle 100 kph
impact with the occupant compartment the load and the deceleration for the camel
impact is higher because of the higher active mass of the camel.
Forty-eight different configurations were simulated and the 90° impact was
identified as worst case scenario for occupant safety.
The evaluation of occupant safety had been conducted on the basis of further
simulations using the finite element simulation model in combination of a multi-
body occupant model. For this, a validated MADYMO occupant model was used.
For this study in a first step it was reviewed, whether the collision with the
vehicle front represents a problem for the occupant. The worst case for accelera-
tion load on the occupants due to an animal impact is, when the vehicle impacts
the animal’s body with full overlap, corresponding to a configuration in which the
animal lies or sits and the whole mass of the animal is effective during the impact
(Fig. 18).
As a reference test the US-NCAP configuration was chosen because it can be
assumed, that for the vehicle design or the design of the restraint systems this con-
figuration is considered. Therefore a FE-simulation with the US-NCAP configu-
ration was made to get the vehicle acceleration, which was used as input for the
MADYMO occupant simulation.
The setup for the MADYMO simulation was a sled with the mounted seat
adopted with respect to the vehicle drawings and a validated HIII dummy
(Fig. 19). Typically restraint systems and characteristics were used, but air-
bag triggering and restraint characteristics were not optimized. Both pulses, the
US-NCAP acceleration pulse and the pulse from the FE simulation against the
Safety Measures for Avoiding or Mitigating the Occupant 75
Fig. 18 FE-Simulation of impact against a block with the animal’s mass, representing the worst
case scenario for occupant acceleration load
Fig. 19 Setup for the multibody-simulation to get the occupant load with respect to the
US-NCAP pulse. Left side FE simulation, right side multibody-simulation
rigid block, representing the animal, were used as input for the MADYMO occu-
pant simulation.
The results of the MADYMO occupant simulation showed that exposure to
occupants are below the level of US-NCAP test, and thus can be assumed that the
acceleration load can be easy handled by the restraint system. The decisive param-
eter for the injury risk in animal accidents is the intrusion in the area of the passen-
ger compartment.
76 W. Sinz et al.
For the definition of the maximum permissible intrusion the space required by
the dummy in the US-NCAP test simulation was used. The path of the outermost
part of the head and the area of the head contact with the air bag form a corridor and
describes a range in which no intrusion may occur. This corridor describes an area
in which the occupant moves with great probability during a crash, and any intru-
sion into this area is thus a risk for occupant injury in case of an animal accident.
Based on the findings of the simulations it was tried to make safety improvements
for the vehicle structure regarding passive safety. Since the intrusions represent the
greatest threat to the occupants, appropriate measures must be made to the vehicle
structure in order to keep the intrusion to an acceptable extent. This illustrates the
above described ‘safety area’, representing the area where no intrusions should occur.
The FE simulations showed, that the mid of the A-pillar is the most criti-
cal area. For an estimation of the force on the A-pillar the assumption can be
made, that an animal body impacts the A-pillar with an impact speed of 72 kph
(20 m/s) and the maximum permissible deformation is 0.2 m, which leads to an
average acceleration of the A-pillar of 100 g. If the effective mass, which means
the mass of the animal which is accelerated, of the animal is for example 300 kg,
this results in an average force of 300 kN. For comparison, the test force for the
federal motor vehicle safety standard roof crush resistance (FMVSS 216a) is for
vehicles with a mass lower than 2,722 kg limited with 80 kN, and this force acts
under a defined angle at the edge of the A-pillar. From the simple example one
can easily see that for the vehicle design no such configuration is considered.
Therefore, it was also decided that for the examined vehicle a solution should be
found that permits an upgrade of the vehicle with the appropriate measure.
One solution is to design a supporting frame for the passenger compartment,
since this brings the one hand, an improvement of the stiffness in this area and
other, to be offered as an upgrade option, if the vehicle should be used over the
whole world. The concept of this ‘cage’ (Fig. 20) consists of a tube-frame of steel
with a diameter of 30 mm and with 3 mm thickness, which should minimize the
intrusions into the passenger compartment. The problems within this solution are
the possible sight obstruction of the tubular frame, because the view field is stand-
ardized, to find an appropriate design in order to implement the load paths in a
Fig. 20 Concept ‘cage’, supporting frame for the front passenger compartment and deformation
of the vehicle at 64 kph impact with visualized safety area
Safety Measures for Avoiding or Mitigating the Occupant 77
meaningful way and to form a supporting structure. Additional problems may arise
with regard to pedestrian protection.
The FE simulations for the ‘cage’ have shown that especially the windshield
and the middle part of the windscreen frame and roof are the vulnerabilities. This
can only be solved by an additional support down to the passenger compartment.
This can be performed either by a split windscreen with an additional beam in the
middle of the windscreen or by an additional carrier in the inner of the vehicle.
Both solutions result in restrictions of other functions such as field of view and
comfort. Depending on the impact speed and the geometric conditions an impact
in the area of the windshield may also occur, and in this case the support beam for
the roof structure causes an improved efficiency.
The calculations showed that by appropriate measures at the structure improve-
ments can be achieved. However, the proof of evidence for real accidents is
limited because there is not enough detailed information about real accidents
available. With the chosen method only relative assessments can be made, that
means it can only be evaluated if the injury risk is increased or not, and only for
the selected load case.
In the area of passive safety the potential is not available to that extent as in active
safety, but there is an additional potential by combining active and passive means.
6 Integrated Safety
As already described, active safety measures have a great potential, such as the com-
bination of object detection and brake assistant. However, regarding the implemen-
tation there are certainly to solve some tasks, such as object recognition, whether the
system is running as a warning or autonomous system, and so on. And also regard-
ing the efficiency there are restrictions, depending on the speed and local conditions
and restrictions it could be possible that the system’s respond to each situation is
ambiguous. The requirements for sensors have been already discussed. In addition,
however, advantages arise from the combination of active and passive measures,
for example if the accident could not be prevented the active measures could lead
to a reduction in collision speed or improvements in collision configuration and in
addition the passive measures will reduce the severity of the accident. These pas-
sive measures could be adopting structure profiles, which are blasted in case of an
impact, or pre-crash airbags. Also seats, which are moved in case of an impact, to
bring the passengers out of the possible deformation area, are conceivable.
7 Results
There are only few detailed statistical data regarding accidents with large animals
and most of statistical data are not precise enough to make concrete statements
about the accident configuration. Therefore, the worst case scenario was detected by
78 W. Sinz et al.
References
1. Bruinderink, G., Geert, W.T.A., Hazebroek E.: Ungulate traffic collisions in Europe. Conserv.
Bio. 10, 1059−1067 (1996)
2. Brockmann, S., Wildunfälle: Untersuchungsgebiet Oberbergischer Kreis NRW;
Unfallforschung der Versicherer. GDV, Presseforum, 25 Apr 2007
3. Björnstig, U., Eriksson, A., Thorson, J., Bylund, P.O.: Collisions with passenger cars and
moose, Sweden. Am J Public Health, 76(4), 460−462 (1986)
Safety Measures for Avoiding or Mitigating the Occupant 79
4. Sullivan, J.M.: Relationship between Lighting and Animal-Vehicle Collisions, The University
of Michigan, Transport Research Institute, Report No. UMTRI-2009-35, Oct (2009)
5. Hedlund ,J.H., Curtis, P.D., Curtis G., Williams A.F.: Methods to Reduce Traffic Crashes
Involving Deer: What Works and What Does Not, Insurance Institute for Highway Safety,
Oct (2003)
6. Al-Ghamdi, A.S., AlGadhi, S.A.: Warning signs as countermeasures to camel-vehicle colli-
sions in Saudi Arabia. Accid. Anal. Prev. 36, 749–760 (2004)
7. Attewell, R., Glase, K., Bull Bars and Road Trauma; Road Safety Report CR200;
Department of Transport and Regional Services, Australian Transport Safety Bureau, Dec
(2000)
8. Tomasch, E., Steffan, H.: ZEDATU – Zentrale Datenbank tödlicher Unfälle in Österreich
– A Central Database of Fatalities in Austria; International Conference, ESAR - Expert
Symposium on Accident Research, (2006)
9. Tomasch, E., Hoschopf, H., Sinz, W., Schretter, N.: Methodik zur automatisierten
Effektivitätsbewertung von Fahrassistenzsystemen am Beispiel von Heckkollisionen. Crash.
Tech. (2012)
10. Sinz, W., Hoschopf, H., Kirschbichler, S., Tomasch, E., Darwish, S.M.H., Almodimigh, M.,
Steffan, H.: Methods to prevent or mitigate accidents with large animals. Characterization
and Development of Biosystems and Biomaterials. ISBN 978-3-643-31469-8. Springer,
Berlin, (2013)
11. Fredin, S.: The Evolution of Passive Safety – How Active Safety may enable. VDI. Ber.
2013, 3−8 (2007)
12. Bishop, R.: A Survey of Intelligent Vehicle Applications Worldwide. In: Proceedings of the
IEEE Intelligent Vehicle Symposium 2000, Dearborn, 3−5 Oct 2000
13. DSD.: PC-Crash Manual. Linz, (2010)
Mechanical Evaluation of Microimplants
for Dental Surgery
1 Introduction
In recent times, the maxillofacial surgery to solve facial fractures has been a chal-
lenge, because of the rehabilitation of these fractures carries the function and the
aesthetics. The facial fractures on the jaw reside in the third place and the subcon-
dylar fractures about 18 % of this third place [1].
Some studies establish the frequency between male and female, as it is shown
in Fig. 1, [2].
As the statistics report for Mexico, genetical problems and the bad habit to
eating and chewing food contributes to injure the jaw in the propagation of some
fractures and decreases the stability around the root of the involved tooth.
An alternative solution is the splinting with microimplants, because they rep-
resent a non-invasive method in a required surgery. The fixation of the jaw is a
requirement to reduce the propagation of the fractures which are present as a con-
sequence of the deficiency of the jaw.
Fig. 2 Maxillomandibular
fixation screws in the
intersection between the
adhered and free gum
Mechanical Evaluation of Microimplants for Dental Surgery 83
The microimplants are used as a new concept for temporary skeletal anchorage
in orthodontic treatment, e.g. for active tooth movement or passive stabilization.
Its special overall design makes it function as a modern system, because of the
wide usage of sterile packaged mini-screws [3] (Fig. 2).
Some advantages of the microimplant system are:
• Enables simpler, more effective orthodontic treatment
• Can replace headgear treatment
• Designed to conserve teeth, avoiding extractions
• Only partial banding required
• Optimized use of time and materials
• It is well tolerated by the patient
• The pins are not visible during treatment [4].
Fig. 3 Constraining points
on the maxillary and jaw
bone to avoid the oral cavity
84 J. A. Beltrán-Fernández et al.
Fig. 4 Upper and lower fixation on the 1st molar with the microimplants to avoid the oral cavity
TENSILE LOADING
Shimadzu testing machine
Tensile loading (3 N)
1 mm/min
FIXSORB-MX Microimplant
constraining the oral cavity
Length of the wire = 25 mm
3 Experimental Analysis
Mechanical tensile testings were applied under the following loading conditions:
– Upper jaw movement—1 mm/min
– Constant loading application during the testing to 100 gf
Mechanical Evaluation of Microimplants for Dental Surgery 85
0.5
-3
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6
Displacement (mm)
4 Discussion of Results
Although the maximum displacement on the elastic zone is about 0.75 mm, the
mechanical behavior of the wire fulfils the expectations of the maxillofacial sur-
geon. It is necessary to avoid the oral cavity. For this, the mechanical response
of the microimplants and the wire is a very important requirement. In words of
the medics, the principal problem for the subcondylar fracture treatment con-
sists on not affecting the root of the tooth and on decreasing the headache of the
patient when a micromplant has been fixed. This is the principal interest that they
expressed and the need to know the mechanical behavior of all the arrangement
is to find a correct constraint to the rehabilitation of the jaw or maxillary when a
fracture is present.
5 Conclusions
Acknowledgments The authors kindly acknowledge the grants awarded by the National
Council for Science and Technology, CONACYT, Instituto de Ciencia y Tecnología del
Distrito Federal (ICyTDF), the National Polytechnic Institute (SIP20091599, SIP20100496 and
SIP20113474) and the support given to this project by the Hospital 1º de Octubre - ISSSTE.
References
1. Rodríguez, R., Cruz, B., Álvarez, B.: Fractura subcondílea. Presentación de un caso.
Mediciego (2011)
2. Zillo, M.: Epidemiology of mandibular fractures treated in a Brazilian level I Trauma Public
Hospital in the city of São Paulo, Brazil. J Braz. Dent. 17-3, Ribeirão Preto (2006)
3. Sánchez, A.: Ferulización de dientes pilares de prótesis parciales removibles a extensión distal
retenida por aditamentos. 42-3 Home (2004)
4. Aydinlik, E., Dayangac, B., Celik, E.: Effect of splinting on abument tooth movement. J.
Prosthet Dent. 49, 447–480 (1983)
Biomechanical Prosthesis Design
of an Orbicular Cranial Cavity
Abstract In this chapter, the design and manufacture of a new prosthesis for an
ocular cavity is reported. The required models were obtained with a computed
axial tomography (CAT) in conjunction with computed aided design (CAD)
and engineering codes. Three hundred and three[p' tomography slices were
taken in digital format (DICOM). Each slice was processed with Scan IP and
Pro/ENGINEER. The main objective of this study is to create a digital model,
which can be fixed as a cranial implant, by using of a quick prototyping printer
system (dust). This prosthesis was implanted on a patient of the Hospital of the
Institute of Security and Social Services for State Workers of the State (ISSSTE).
For this purpose, craniumplasty was the surgery technique that was followed. This
technological development was based on the obtainment of quick prototypes,
which is applied in orthopedics. The clinical case discussed in this chapter is
related to a 24-year-old individual. He was treated in the area of maxillofacial and
plastic surgery. The traumatism was derived from an impact of bullet on the ocular
right region of the cranium. Calcium hydroxyapatite, which is highly biocompat-
ible, was used for this purpose. It is very important to mention that this procedure
avoided the employment of bone tissue taken from the ribs of the cranial region.
As a result of this research, a 3D personalized orbicular cranium model could be
presented in order to be fixed in the patient.
1 Introduction
Fig. 2 a Cervical porcine instrumented specimen in a Tomographer. b 3-D rebuilded cervical
porcine model. c Digitalized cervical porcine model [2]
not clearly defined. Afterward, an alternative methodology was used in the analysis
of cervical and lumbar vertebrae 3-D (Computed axial tomography (CAT). For this
purpose, Digital Imaging and Communications Format files (DICOM) were used in
conjunction with a 3-D helical scanning systems (Fig. 2). The cortical and cancel-
lous tissues could be seen on the tomographic slice and the geometrical characteris-
tics of the model were much close with those of the vertebral bodies [1].
Under this methodology, a precise bone reconstruction could be done, taking
into account each computed tomography of the patient and the specialized soft-
ware: SCAN IP, Geomagics, Pro-Engineer and Unigraphics.
Taking into account the last experiences, the motivation of this research was
based on the reconstruction of an orbicular cavity prosthesis. For this purpose,
advanced computational tools and a 3-D quick prototyping system was used. It
allowed solving the consequences of the trauma of the patient by the use of mixed
material, synthetic and natural bone [2].
Great advantages of the biomodelling using CAT, is that a quick prototype can be
reproduced accordingly to the anthropometric features and requirements of the patient.
Based on the last experience, a preliminary model was developed. In this way, the
relevant details were obtained before the invasive surgeries. With this information,
a strategy was developed for the solution of this case.
90 J. A. Beltrán-Fernández et al.
(a) (b)
Fig. 4 a TAC of the patien considering an explosive detail of the orbicular cranial. b Fractured
and damaged zone
Biomechanical Prosthesis Design of an Orbicular Cranial Cavity 91
Fig. 5 a Lateral view of the delimited drawing zone, b Top view of the delimites drawing zone
Each computed tomography (CT) image was defined using the paint tool of
SCAN IP, taking into account the boundary of the cortical bone. As can be seen in
Fig. 5, the limits of the mask are in accordance with the density of the picture. It is
delimited by the balance between the brightness and contrast of the tomography.
Once the boundary was closed, the rest of the area was filled automatically. This
is an advantage, as a precisely profile of each section and the cranium are obtained
(Fig. 6).
A stereolithography format (STL) file was obtained and the full 3-D model was
developed (Fig. 7). It was processed as parametric format (faceted model) in dif-
ferent computed codes as Pro-Engineer 4.0 Wildfire, and UGS-NX 7.5.
92 J. A. Beltrán-Fernández et al.
Fig. 7 Quick prototypes printed in a 3-D system, using calcium dust and hydroxyapatite
Fig. 8 Damaged status of the orbicular cranial cavity (left) and the first propose of the fixing
sane cavity (right)
Because of the fact that the surgeons required an initial assessment in non inva-
sive way, two quick prototypes were printed. The damaged area and the first pro-
pose of the implant are shown in Fig. 8.
A quick prototyping system (ZPrinter 650–Z Corp) with a build size of
10 × 15 × 8 in. (254 × 381 × 203 mm) and a layer thickness of 0.0035–0.004
inches (0.089–0.102 mm) was used for the manufacturing of this prototype. This
quick prototyping system functions as a conventional 2-D inkjet printer. The mix-
ture of hydroxyopatite and resin was distributed over a plane surface. However,
this printing system applies a polymer resin taking into account the drawn profiles
of each tomographic slice. Also, there is a vertical displacement, which is perpen-
dicular to the principal plane surface. This could not be done with 2-D inkjet sys-
tems (Fig. 9).
It is very common to use titanium plates or medical steels for the most of sur-
gery procedure solutions or when a new prosthesis is required. However, higher
costs are frequent. In contrast, the replacements are carried on with other bony tis-
sues, which are taken from other bones of the patients. For the problem at hand, it
was used hydroxyapatite to develop the required the required prosthesis. The main
reason of its inclusion is focused on its biocompatibility. This material can be inte-
grated with the bony structure after 6 months of the surgical operation.
Biomechanical Prosthesis Design of an Orbicular Cranial Cavity 93
Fig. 9 Final 3-D rebuilding and digital model considering the correction of the damaged zone
3 Results
Actually, the prosthesis has had a good performance and FEM analysis can be
done, however, for this chapter this analysis will not yet be carried out. Also, it
is possible to mention some interesting points, in order to get an idea of the com-
plexity of the geometry and the requirements to be processed. 6,58,453 facets
were generated. It allows establishing an analysis of the model in a FEM code.
However, high computational resources are required.
4 Conclusions
Modern engineering evaluation of medical solutions are not only required in the
orthopedics but also in the maxillofacial area. In all these cases, the knowledge
of the biomechanical behavior of the involved bony structures is required [3]. For
this case, the cranial cavity contributes in the protection of the structural integrity
of the ocular cornea. The analysis of the damaged zone, before the surgery, did not
assure the correct functionality of the involved muscles. The low cost of this pro-
cedure and its simplicity contributes in the solution of similar cases. This is one of
the most recommendable methodologies that medics and hospitals have approved.
It has to keep in mind that this procedure depends on the conditions of the dam-
aged area and the correct alignment of the patient during the tomography session.
Furthermore, the processing of the models with DICOM files is a good alternative.
Its main advantages are: it is non invasive technique and geometrical parameters
are obtained directly.
94 J. A. Beltrán-Fernández et al.
Acknowledgments The authors kindly acknowledge the grants awarded by the National
Council for Science and Technology, CONACYT, Instituto de Ciencia y Tecnología del
Distrito Federal (ICyTDF), the National Polytechnic Institute (SIP20091599, SIP20100496 and
SIP20113474) and the support given to this project by the Hospital 1º de Octubre - ISSSTE.
References
1. Beltrán, J. et al.: Assessment of the structural integrity of c3–c5 cervical porcine vertebrae
model based on 2-D classic cad, 3-D scanner and 3-D computed tomography. In: Öchsner A., da
Silva L.F. M., Altenbach H. (eds.) Analysis and Design of Biological Materials and Structures,
vol. 1/2010– vol 20, (2010) Springer, Berlin ISBN: 978-3-642-22130-9
2. Beltrán, J. : Manufactura de prótesis de fosa ocular en impresora de polvos 3D, Selección
Gaceta IPN, Número 36, Año IV, vol. 4, 31 de Octubre de 2011, ISSN: 0061–3848
3. Beltrán, J. et al: Biomechanical characterization of a cervical corporectomy using porcine
specimens, following an experimental approach. J. Key Eng. Mater. 478, 103–108 (2010)
ISSN: 1662–9795
Biomechanical and Control Design
of a Prototype for Muscular Re-Training
of a Forearm and a Wrist
Abstract In this chapter, the design and control of a forearm and wrist re-trainer
machine is reported. This equipment was designed using the anthropometric
standards of Mexican individuals, who are in the range between 18 and 65 years
old. The main objective of this retraining device is to recover the muscular resid-
ual capacity of individuals, who have suffered a disability in the forearm and/
or the wrist after an accident or as result of a genetic heritage. Quality Function
Deployment was used in order to fulfill the requirements of patients, therapists and
orthopedists. The logic of control was developed considering a user interface and
the workflow programing. The analysis of the performance of the first prototype
suggested the improvement of the three areas: the mechanism, the control system
and acquisition system. The mechanical virtual design and the Human Machine
Interface (HMI) of the prototype were approved by the Institute of Health of the
State of Mexico. A final evaluation of the medical regulatory body is awaited.
1 Introduction
The upper extremities are very important in the daily activities. In this case, the
forearm and wrist play an important role. These musculoskeletal components
allow the manipulation of diverse objects. In occasions, due to accidents or genetic
diseases, these elements are disabled or become too weak for the development of
common activities. In these cases, rehabilitation is required in order to improve the
quality of life of patients.
The use of robotic equipment for rehabilitation is not a new idea, as an exam-
ple, the following patent [1] can be mentioned. Nowadays, the range of applica-
tions has increased [2]. Most of the movements developed by superior extremities
are complex. However, robotic devices allow the characterization of such move-
ments [3]. Moreover, these analyses have also been applied to lower extremities
[4]. These experiences have been applied in the optimization of the rehabilita-
tion process [5] or adjust control techniques have also been followed in order to
improve the rehabilitation effect [6].
In order to apply surgical procedures to the wrist, medics have some issues [7].
A common postoperative practice is the immobilization of the limb. This proce-
dure includes the proximal and distal articulations. It has to kept in mind that fixa-
tion affects muscles and articulations. In order to avoid this harmful effect, robotic
devices are used [8, 9].
Robotic devices are also used in the rehabilitation of patients who have suf-
fered stroke [10, 11] or tetraplegia [12]. In order to establish a comparative level
between muscular tone and the therapies, ultrasound equipment can be used [13].
Another interesting point is that biomedical designs can be combined with compu-
tational techniques, these results in virtual reality equipment [14, 15].
In general terms, biomedical devices are useful for therapist, medics and
patients. However, the kinematics and dynamics of the human body is complex. The
development of rehabilitation equipment, which satisfies all the needs, is a difficult
task without a proper methodology. In this case, the use of the Quality Function
Deployment (QFD) methodology has been very valuable; the qualitative to quan-
titative interpretation of the main needs are briefly resumed. A proper design is
obtained easily. The proposed design, discussed in this chapter, is mainly focused in
three needs: wrist rotation (forearm pronation and supination), flexion and extension
of the forearm and the development of a retraining device with a compact structure.
Biomechanical and Control Design of a Prototype for Muscular Re-Training 97
The forearm is between the elbow and the wrist and contains two long bones: the
radius and the ulna. Ten muscles are inserted in the radius and fifteen in the ulna.
The forearm is capable to develop movements of flexion and extension. Besides,
the wrist is the articulation that allows joining the forearm with the hand. It is
called condyle because it allows the development of the following movements:
flexion, extension, abduction, adduction and circumduction.
A new prototype, which helps in the rehabilitation of the forearm and wrist, is
required. It has to satisfy the needs expressed by medics and therapist. Such equip-
ment has to be mobile and portable. Besides, it will be used for the therapy of
Mexican patients (women and men), whose age are in the range between 18 and
65 years old. The lengths considered for forearm and hand palm are 243 mm and
85 mm, respectively [16]. A pneumatic piston is proposed to control the flexion angle.
The main movements considered in this equipment are flexion, extension, pro-
nation and supination. In general terms, the movements used for rehabilitation
must be aligned with sagittal, coronal and transversal planes. Once these condi-
tions are satisfied, therapist must observe if the movements developed are against
gravity or pro-gravity.
In the development of the control and automation, the flow of process has to be
considered. In general terms, the sequences involved are: the patient is put in the
correct posture, the data (as name, sex, and age, among others) is logged and the
sequence of the therapy is executed. For the development of this work, the authors
also considered the previous experience in the design of a training apparatus for
lower limbs [17].
Table 1 Qualitative requirements
Qualitative need Engineering parameters
Light device Density and mechanical properties of materials
Continuous passive movements Mechanisms, time, power, geometry, anthro-
pometric characteristics, stability, rotation,
flexion and extension.
Rotation of wrist Angular position, rotation velocity and security
parameters
Record of the progress of each patient Human Machine Interface (HMI) and
Programmable Logic Controller (PLC)
Sterilization Type of materials used
Adjustable to various dimensions Mechanisms and patient restrictions
Biomechanical and Control Design of a Prototype for Muscular Re-Training 99
Table 2 Quantitative references
Qualitative needs Engineering reference
Light device 5 kg is the top limit of the weight of the
equipment
Continuous passive movements A mechanism was developed considering the
anthropometrical characteristics of individuals
between 18 and 65 years old. Care was taken
in the consideration of the anthropometric
limits for the forearm and the wrist.
For evaluation of stresses and strains, the range of
mass was considered between 2 and 10 kg
Evaluation of the progress of the patients A Personal Computer is the main interface.
Therefore, data are acquired using a HMI.
Control is applied through a PLC and
H-bridge. Alarms and an emergency stop are
activated when a malfunction is detected or
the limits established are over range
Main mechanic characteristics The rehabilitation device was mainly made
with plates of aluminum 6061 and stainless
steel 306. The mechanism and two gears
were specially designed for wrist rotation. A
pneumatic piston and an electro valve control
the angle of flexion The device is capable to
develop flexion of forearm and wrist rotation
simultaneously. In order to give comfort to a
patient, a neoprene bed is put for the support
of the forearm
Main control characteristics The PLC has six digital inputs, two analogic
input-digital inputs, six digital outputs and
one analogic output. A comprehensive inter-
face is designed in the graphic builder
The QFD gave a lot of information. However, the parameters defined involve a retro-
spective research and a benchmarking analysis. In the first step, diverse alternatives
were evaluated. The objective was the development of a good design. This implies
that minor changes have to be done during the manufacturing process; consequently,
time and money were saved. Figure 1 shows a schematic arrangement of the reha-
bilitation device designed. The main components are illustrated. The main support
for the hand (1) was made of aluminum 6061. The forearm support (2) was made of
a stainless steel plate and covered with a bed of neoprene with strips of Velcro. The
gear mechanism (3) for the hand was made of stainless steel, as well as their com-
ponents. The pneumatic piston (4) allows and controls the movement of flexion. The
gearmotor with an encoder (5) provides a rotatory movement for the hand.
Figure 2 illustrates how it was assembled. The main components are: The
container box (1), covers (2) and transmission shaft (5). They were made with
100 A. Luna-Avilés et al.
Fig. 2 Prototype assembly
Biomechanical and Control Design of a Prototype for Muscular Re-Training 101
stainless steel 306. The main platform (3) was made of aluminum 6061 as well
as the base (6). The hand mechanism (4) has two covers for protection. All the
control devices are inside the rehabilitation equipment. Figure 3 shows the way in
which they are positioned.
Finally, the device was programmed through a Human Machine Interface
(HMI). The rehabilitation parameters, linear and angular velocities and displace-
ments, were controlled. Figure 4 shows schematically the rehabilitation process
and the control diagram is illustrated in Fig. 5.
The prototype uses six digital outputs and four digital inputs. These control sig-
nals are useful to regulate the spin of the encoder in the gearmotor and the linear
displacement of pneumatic piston through the electrovalves.
Fig. 3 Equipment approach
Fig. 4 Schematic process
102 A. Luna-Avilés et al.
Fig. 5 Control diagram
1. When the PB, H and HMI are activated, a pulse is sent in order to set the EV 1.
The purpose is preparing the device to operate.
2. Once all the conditions are satisfied, LD, AD and AC, PB H ST HMI reset the
EV 1 and a pulse is sent to set EV 2 in order to put the device in the position
predefined by the therapist.
3. In this position, the prototype can start the rehabilitation process. The rehabil-
itation device can execute linear or angular displacements, simultaneously or
individually. This is because the H–B and the EV 1 and EV 2 are controlled
separately.
4. An analog comparator is included in order to control angular displacement and
to avoid injuries. Also a PB EB HMI is included for security purposes.
5. Finally, all therapy is recorded in the main controller device (CPU).
Biomechanical and Control Design of a Prototype for Muscular Re-Training 103
The rehabilitation device can be operated easily and is comfortable. Besides, there
is a continuous feedback of the individual rehabilitation and the rehabilitation rou-
tine is maintained. It is possible to add thermo therapy equipment during the con-
tinuous passive movement (CPM).
On the other hand, the main disadvantages are the following. It requires an aux-
iliary equipment (compression air system), which weights 10 kg. Friction between
the skin of the individual and neoprene must be consider in diabetic patients; and
damage could be inflicted to a patient if the device is not used properly.
The rehabilitation machine was evaluated by the health authority. In this case,
it is the Institute of Health of the State of Mexico. The health authority established
that this device could be useful to improve the quality of life of patients. It was
suggested that it could be used in conjunction with other rehabilitation therapies. It
is important to consider that a good balance must be developed between the reha-
bilitation process and the rehabilitation routine. The evaluation that was mentioned
before was applied through a CAD system and a simulation software.
4.1 Improvements
Once the first functional prototype was manufactured and tested, several condi-
tions were improved due to recommendations of the users and the manufacturing
department. The main modifications were made to: the mechanical structure, the
mechanism which allows the complete vertical movement and the HMI control
system. A schematic comparison is show in Fig. 6.
The principal reasons of these modifications are the following: (a) additional
support to the arm is required and the rigidity of the device has to be improved, (b)
the dimensional restrictions of the gearmotor have to be optimized, (c) the shaft
deformation has to be eliminated, (d) the mechanical assembly must be improved
and (e) the obstructions during vertical displacement must be eliminated. Figure 7
shows the two main stages during a vertical operation.
Regarding the control system, the HMI was designed to control the electro
valves and the gearmotor with an encoder. The first control gives the possibility to
regulate the vertical displacement (pneumatic actuator). The mechanical character-
ization had been programed with a PLC. The analysis of the results of several tests
suggested the development of a new code. The modifications included: (a) a check
of the initial status for the operation of the device; (b) If a power failure or an
emergency take place, the mechanical device must return to home (cero machine
coordinates); (c) The data of each patient must be downloaded or uploaded to/
from excel or access codes; and (d) The therapy execution has to be interpreted
with comprehensive graphs.
In the first two cases, the initial status was considered. This step was developed
in automatic way with a new code. Regarding the final point, several medics and
therapist have the records of their patients in excel and/or in access data bases.
Therefore, time is saved when the information is handled to/from such data bases
through a hyperlink in the HMI. The variables considered for this purpose were:
(1) name, (2) age, (3) sex, (4) treatment, (5) sickness and (6) information of previ-
ous sessions.
Finally, the functional prototype must be validated by a medical regulatory
body. Therefore, it is important to keep a comprehensible statistic. The vari-
ables selected for this purpose were: (a) initial and final vertical displacement,
(b) initial and final angular displacement, (c) quantity of treatments, (d) parallel
therapies, such as thermotherapy and (e) muscular tone oriented to atrophy or
hypertrophy.
5 Conclusions
This chapter illustrates the procedure that was followed in the development of a
functional prototype called “Muscle re-educator for rehabilitation of a forearm
and wrist”. The conceptual design and the geometric dimensions were obtained
through the rehabilitation requirements and anthropometric characteristics of
patients. Adjustments for different sizes were limited in the range of 12 mm. Also,
there are some restrictions on the rotation and flexion angle. These are in accord-
ance with the anthropometric characteristics of the Mexican individuals.
All physical parameters involved (velocity and displacement) have been estab-
lished, in such way that the structural integrity of the musculoskeletal system is
maintained during the rehabilitation period.
The initial prototype proved to be efficient. At the same time, a clinical proto-
col was developed with the help of the health authority, in order to establish “the
standard gold.” A balance between the clinical treatment and the rehabilitation
program. Once the prototype has been validated, a re-design process took with
the final remarks. The experience obtained with initial prototype was used in the
improvement of the device. These were done in the mechanism design, the control
system and the data acquisition system.
Acknowledgments The authors kindly acknowledge the support given to this Project by
TESCo, PROMEP, ISEM, National Polytechnic Institute and the Institute of Science and
Technology of the Federal District.
References
1. Reddy, et al.: Arm rehabilitation and testing device. US Patent 5,170,777 (1992)
2. Masia, L., Krebs, H.I., Cappa, P., Hogan, N.: Design and characterization of hand module for
whole-arm rehabilitation following stroke. IEEE ASME Trans. Mech. 12(4), 399–407 (2007)
3. van Mourik, A.M., Daffertshofer, A., Beek, P.J.: Extracting global and local dynamics from
the stochastics of rhythmic forearm movements. J. Mot. Behav. 40(3), 214–231 (2008)
106 A. Luna-Avilés et al.
4. Seddiki, L., Guelton, K., Zaytoon, J.: Concept and Takagi–Sugeno descriptor tracking con-
troller design of a closed muscular chain lower-limb rehabilitation device. IET Control T.
App. 4(8), 1407–1420 (2010)
5. Ellis, M.D., Sukal-Moulton, T., Dewald, J.P.A.: Progressive shoulder abduction loading is
a crucial element of arm rehabilitation in chronic stroke. Neurore. N. Rep. 23(8), 862–869
(2009)
6. Dumitriu, A., Barbu, D. M.: Use of inertial sensors in lower-limb rehabilitation device. Proc
20th Int DAAAM 20(1): 1209–1210 (2009)
7. Gaulke, R., Abdulkareem, M., O’Loughlin, P.F., Oszwald, M., Probst, C., Hildebrand, F.,
Krettek, C.: First clinical experience with a novel forearm boom. Tech. Health Care 18, 317–
324 (2010)
8. Berner, S.H., Willis, F.B.: Dynamic splinting in wrist extension following distal radius frac-
tures. J. Orth. Surg. Res. 5, 1–4 (2010)
9. Matsumura, M., Ueda, C., Shiroishi, K., et al.: Low-volume muscular endurance and strength
training during 3-week forearm immobilization was effective in preventing functional dete-
rioration. Dyn. Med. 7, 1–8 (2008)
10. Kan, P., Huq, R., Hoey, J., Goetschalckx, R., Mihailidis, A.: The development of an adaptive
upper-limb stroke rehabilitation robotic system. J. Neuroeng. Rehabil. 8(33), 1–18 (2011)
11. Mayhew, D., Bachrach, B., Zev Rymer, W., Beer, R. F.: Development of the MACARM – a
novel cable robot for upper limb neurorehabilitation. Proc. 2005 IEEE 9th ICRR 299–302
(2005)
12. Glinsky, J.A., Harvey, L.A., Ben, M.: A new clinical device for measuring wrist strength in
people with tetraplegia. Physiother. Theory Pract. 26(5), 342–346 (2010)
13. Guo, J.Y., Zheng, Y.P., Huang, Q.H., Chen, X.: Dynamic monitoring of forearm muscles
using one-dimensional sonomyography system. J. Rehabil. Res. Dev. 45(1), 187–196 (2008)
14. Maniya, J.B., Pathak, P.M., Mishra, B.K.: Design and development of virtual objects to be
used with haptic device for motor rehabilitation. J. Softw. Eng. Appl. 3, 990–997 (2010)
15. Subramanian, S., Knaut, L.A., Beaudoin, C., McFadyen, B.J., Feldman, A.G., Levin, M.F.:
Virtual reality environments for post-stroke arm rehabilitation. J. Neuroeng. Rehabil. (2007).
doi:10.1186/1743-0003-4-20
16. Avila-Chaurand, R., Prado-León, L R., González-Muñoz, E. L.: Dimensiones antropomé-
tricas de población latinoamericana. Universidad de Guadalajara QADD (2001)
17. Luna-Avilés, A., Hernández-Gómez, L.H., Beltrán-Fernández, J.A., Urriolagoitia-Calderón,
G., Gómez-Michel, D., Martínez-Espinoza, D.M., Villegas-Acosta, E., Urriolagoitia-Sosa,
G.: Using quality function deployment methodology to translate qualitative to quantitative
requirements in the design a of a knee re-trainer. In: Öchsner, A. (ed.) Characterization and
development of biosystems and biomaterials. Springer-Verlag, Berlin (2013)
Optimization of the Design of a Four Bar
Mechanism for a Lower Limb Prosthesis
Using the Taboo Search Algorithm
Abstract In this chapter, the optimization of the design of a four bar mechanism
used in polycentric prosthesis with voluntary control is reported. This prosthe-
sis has been used by a male, whose left leg was amputated above the knee. He is
34 years old, weighs 78 kg and is 1.75 m tall. One of the objectives of this opti-
mization was to fulfill the requirements of his anthropometric characteristics. The
taboo search algorithm was used for this purpose. In this case, the lengths of the
links were determined, following an inverse analysis. The objective function was
the minimization of the error between the target trajectory of the instantaneous
center of rotation (ICR) and the path followed by ICR of the four bar mecha-
nism. This curve is very important because it is related with the kinematic and the
forces that are developed in the gait cycle. Therefore, it is expected that the ampu-
tee individual develop a natural gait cycle with this prosthesis. In the final stage,
1,900 iterations were carried out and the lengths of the bars were augmented by
0.01 mm. The lengths of the links of the optimized mechanism are the following:
Bar 1 = 43 mm; Bar 2 = 55.5 mm; Bar 3 = 59 mm and Bar 4 = 29 mm. With this
information, the prosthesis was manufactured and adapted to the patient.
1 Introduction
The kinematic analysis of a knee joint is complex. Gunston [1] showed that its
movements combine rotation, oscillation and translation. They have a strong
relation with the forces that take place during the walking cycle. As a result, the
instantaneous center of rotation (ICR) of the knee joints moves along a centrode
trajectory. In Fig. 1, it is illustrated this trajectory and the curve which is devel-
oped during the first 10º of flexion. Besides, it is common to use a four bar mecha-
nism in this sort of prosthesis. Radcliffe [2] established that the ICR should be
located, in such way that an individual with a transfemoral amputee would have
Fig. 1 Mechanism of four
bar used in a polycentric
prosthesis with voluntary
control. a Total trajectory
of instantaneous center of
rotation. b Trajectory of
the instantaneous center of
rotation during the first 10º of
the flexion of the leg
Optimization of the Design of a Four Bar Mechanism 109
a full command of the prosthesis. Accordingly, the main idea is that the trajectory
described by ICR of a lower limb prosthesis with a four bar mechanism, must be
very similar to the trajectory of the moving centrode, which is generated by the
knee joint during a natural gait cycle.
For this purpose, an optimization of the polycentric mechanism of the knee
prosthesis is required. The appropriate dimensions of the links are determined. As
a result, the different phases developed in a normal walking cycle are reproduced
and a good stability is provided. This approach was discussed in [3, 4].
There are several procedures for the optimization of the design of knee pros-
thesis with a four bar mechanism. An alternative is the Taboo search algorithm
[5, 6]. The word taboo can be defined as “things that are untouchable because
they are sacred.” In the case of the taboo search algorithm, it can be said that the
solution is based on searching by cycles in the space of solution. However, it is
required to seek in different points of such space. For this reason this algorithm
prohibits or penalizes to the points in the solution space which have been previ-
ously visited. Hence, these conditions are defines as taboos.
The taboo search algorithm is a heuristic procedure for solving optimization
problems. One of its main characteristics is the avoidance of local optima. This
algorithm implements flexible structures of memory. Thus, the process of search
information is agile. The intermediate results are used in obtaining new solutions
in which the parameters of interest are optimized. Also, conditions for strategically
constraining and freeing the search process can be considered [7]. Due to the flex-
ibility of this approach, customized mechanisms can be obtained easily, consider-
ing the anthropometry of each individual.
In the optimal design of four-bar linkages for path-generation purposes, the trajec-
tory of the coupler must track the precision points. Thus, the problem is to establish
the dimensions of the links in order to follow the required path, Khorshidi et al. [8].
Smaili et al. [9] made the synthesis of a four bar mechanism with the taboo-gradi-
ent algorithm. They compared the results with those obtained with other optimization
procedures. They concluded that the taboo search methodology got the best solution.
Regarding this chapter, the design of a four-bar knee mechanism with volun-
tary control used in a polycentric prosthesis was optimized. In order to obtain the
dimensions of the links of the mechanism, the taboo search algorithm was used.
A male, who is 34 years old, was amputated above the knee. He weighs 78 kg and is
1.75 m tall. The patient has been using a transfemoral prosthesis since he was seven
years old. As he is a young patient, a four-bar knee prosthesis with voluntary control
for his left leg was advised. He was treated in the Department of Rehabilitation of the
1º de Octubre Hospital in Mexico City. An informed consent from the patient was
obtained. For this purpose, the optimization of the design of a polycentric prosthesis
was required, in order to fulfill the requirements of his anthropometric characteristics.
110 J. J. Muñoz-César et al.
In order to obtain a good design, a target trajectory on the ICR on the sagittal
plane was proposed. This curve is related with kinematic and the forces that are
developed during the gait cycle. Therefore, the path followed by the ICR of the
four bar mechanism proposed must reproduce such trajectory. In this condition the
amputee individual has a good stability and control in a gait cycle. This design is
based on the stability analysis proposed by Radcliffe [10], which is appropriate for
a four bar polycentric mechanism. In this analysis, the flexion and extension of the
knee were considered
In this inverse problem, the appropriate dimensions of the bars were deter-
mined. This is an optimization problem and it can be done with different proce-
dures. However, the taboo search algorithm was selected, as it was considered
adequate for this purpose.
Four-bar linkage knee mechanisms have been used in trans femoral prostheses.
However, they are only suitable for certain patients. Radcliffe [2, 11, 12] estab-
lished the localization of the ICR, in order to obtain a suitable polycentric mecha-
nism and a transfemoral amputee has a full command of these sort of prostheses.
The motion of the human knee joint is more complex than a single rotation around
a fixed axis. It is the result of three combined motions that allow a controlled gait
cycle, as shown in Fig. 2. These are: (1) rotation from 0° to 10°, (2) oscillation
from 10° to 20° and (3) translation from 20° to 120°. The ICR of the whole sys-
tem moves along a specific trajectory, from the posterior to the anterior side of the
thigh, where the socket is commonly attached.
Radcliffe [12] explained the line load is used to assemble the components of
the prosthesis, in such way that the prosthesis provides stability in weight bear-
ing. However, there are some small changes during the support in the gait cycle. In
order to maintain the equilibrium, there is a “dynamic alignment”. A fine tuning of
the prosthesis is essential for the requirements of each individual amputee.
In accordance with [11], the main aspects, which are required, to obtain a good
stability are: (1) The length and strength of the residual limb; (2) the fitting of the
stump-socket interface; (3) the relative position of the hip joint with respect to the
knee and ankle joints of the prosthesis (alignment geometry), and (4) the func-
tional characteristics of the knee and foot–ankle mechanisms should be incorpo-
rated into the prosthesis.
In accordance with the ideas mentioned above, there are three different types of
four-bar linkage mechanisms used in lower limb prosthesis [10]: (1) the four-bar
mechanism with elevated ICR, (2) the hyper-stabilized four-bar mechanism, and
(3) the voluntary control four-bar mechanism.
The four-bar linkage with an elevated ICR is advisable for geriatric amputees
or those with limited ability to control the stability, using the residual hip function
on the amputated side (Fig. 3). Its anterior link is long and the posterior is short.
The links of this mechanism have been designed to give extreme stability at heel
contact. The ICR for full extension of the knee is located behind the load line at
heel contact. The knee is forced into extension and is cinematically locked during
the extension; no hip extension moment exerted by the amputee is required. The
Fig. 3 Three different types of four-bar linkage mechanism used in lower limb prosthesis
112 J. J. Muñoz-César et al.
amputee develops a hip flexion moment at push off. In this way, the individual
is capable to shift the load line behind the ICR, as it is required to initiate a knee
flexion. The knee flexion is initiated with a minimal effort by the amputee.
The elevated ICR must move downward very rapidly in a knee flexion, when
the individual sits down. This sudden shift of ICR does not allow the amputee to
maintain the control of the weight-bearing knee.
It should be noted that the position of the ICR in full extension is very sensitive to
small angle changes. This feature can be used as a means of adjustment of the knee
stability in full extension by small changes in the position of the knee extension stop.
The arrangement of the hyper-stabilized four-bar knee mechanism is very similar
to the four-bar mechanism with elevated ICR. However, its kinematic behavior is dra-
matically different when small changes in the dimensions of the mechanism are intro-
duced. The ICR is located well behind the load line (Fig. 3). No hip extension moment
is required. At push off, even with the maximum possible hip flexion moment exerted
by the amputee, the ICR is still behind the load line. A knee flexion cannot be initi-
ated, while the prosthesis is weight-bearing. This mechanism is advisable for geriatric
or less active amputees, who require an equivalent of a knee lock in the stance phase.
The voluntary control of a four-bar knee (Fig. 3) has been designed to give to
the amputee the ability to control knee stability at both heel contact and push off.
Also, it has a complete control of knee stability over a limited range of knee flexion.
The control of the motion and its stability depend on the physical capabilities of the
individual. The knee is moved voluntarily to a stable position in full extension. The
reactions of the individual become almost involuntary, as experience is gained.
The voluntary control of the stance phase stability plays an important role over
the first oluntary control of the stance phase stability plays an important role over
the first 10° of flexion, from a position of full extension. The path of the ICR does
not begin at an extremely elevated and posterior position in full extension, as in
the other two types of four-bar mechanisms discussed above. The ICR is located
100 mm above and 6 mm behind of the reference line [10]. From this point, the
ICR can move smoothly. It will go forward and downward, as the flexion angles
of the knee increase. Along this trajectory, the stability is always assured. It is per-
ceived by the patient as a “gliding motion” of the knee block over the shank, with
a controlled flexion and extension of the knee. This controls the stability, while the
patient walks on rough ground, sloping surfaces, or stair descent. This point has
been considered in this optimization. Figure 4 illustrates the estimated trajectory
of the ICR for the problem at hand.
Hobson and Torfason [4] have proposed the following methodology in order to
find the desired trajectory of the displacement of the ICR. The frame linkage is
located at shank of the prosthesis. This is the bar 1 (Fig. 5). Bar 3 is the coupler
Optimization of the Design of a Four Bar Mechanism 113
Fig. 4 Trajectory of the
target trajectory of ICR and
initial four bar mechanism
proposed
Fig. 5 Instantaneous center
of rotation of the proposed
four bar mechanism
and is located in the socket. The crank pins are located at both ends of Bar 1, OA
and OB. The first and second cranks are bar 2 and bar 4, respectively. The lengths
of each bar are a1, a2, a3 and a4. The angles of each bar are θn. They are measured
anticlockwise.
114 J. J. Muñoz-César et al.
The input is defined by the angle of the coupler link (θ3). Equations (3) and (4)
are written with the dependent variable θ2, as
a2 cos θ2 = a4 cos θ4 + C1 (5)
a2 sen θ2 = a4 sen θ4 + C2 (6)
where
C1 = x0B − x0A − a3 cos θ3 (7)
C2 = y0B − y0A − a3 cos θ3 (8)
Both are constants for different values of θ3. Equations (5) and (6) are squared
and added
a22 = a24 + C21 + C22 + 2C1 a4 cos θ4 + 2C2 a4 sen θ4 (9)
−1 A ± A2 + B2 - C2
θ4 = 2tan (16)
B+C
θ4 has a couple of solutions, one in which the radical is positive and other
negative. The sign is selected in accordance with the configuration of the
mechanism.
Up to this point, θ3 is proposed and θ4 is calculated. Following this procedure,
the coordinates of A and B can be found.
yA - y0A
θ2 = tan−1 (21)
xA - x0A
The ICR of the four bar linkage can be calculated from the coordinates of the
centers of rotation of links 2 and 4 and their angles.
The coordinates of the coupling point can be estimated with the distance ac. It
is between A with an angle φ and the line AB. In accordance with this,
θC = θ3 + ∅ (24)
xC = xA + aC cos θC (25)
yC = yA + aC sen θC (26)
Figure 4 shows the trajectory of the moving centrode, which was calculated
with the procedure described above. For this purpose, the four bar mechanism and
the displacement of the ICR are shown. This situation takes place when the leg is
bent from 0º to 90º.
The dimensions proposed initially guarantee the stability and a good weight-
bearing. They are in accordance with the stability criteria for a control voluntary
mechanism. In this way the individual can control the movements during the leg
flexion. Besides, lower loads of the hip are required in order to keep the exten-
sion of the leg and start its flexion. From these conditions, the optimization was
done.
116 J. J. Muñoz-César et al.
Is the
No objective Yes Final
function solution
satisfied?
fact, this is a set of rules. Therefore, the solutions of x inside the neighborhood
are length of the links. For the problem at hand, the initial lengths, which were
proposed for each bar, were the following: Bar 1 = 43 mm; Bar 2 = 56.5 mm; Bar
3 = 30 mm and Bar 4 = 60 mm. With this information, the following matrix was
developed.
B1 43 B2 56.5
B330 B460
This shows the maximum size of the neighborhood, which can be obtained. It
is supposed that the neighborhood is formed by one level of action around the size
of a link. Each action has a position (i,j) in the matrix, where i is the row and j the
column. In this way, each action has a neighborhood of fixed size. The length of
the bars can take values which are in the dominium of the initial matrix, because
this is a random search.
5.3 Objective Function
The objective function calculates the parameters of decision and the limits of the
problem are the restrictions of the solutions. In other words, the limits establish
the values which cannot be evaluated by the objective function. Therefore, those
limits must be in terms of the decision parameters. In certain situations, those lim-
its are considered in terms of equalities or inequalities.
In the problem at hand, the objective function tries to find the best dimensions
of the prosthesis, in such way that a good convergence is obtained with the moving
centrode. Therefore, it is minimized the sum of the quadratic error during the first
10° of the knee flexion of the moving centrodes between the target ICR trajectory
and the one of the proposed prosthesis.
f (x) = e2
5.4 Taboo Structure
The taboo search is based on the structures of the short term memory. It logs the
recent solutions, which are related with the dimensions of the bars. Every time
that a solution is obtained, it is introduced in a list. If the dimensions of the bar are
Optimization of the Design of a Four Bar Mechanism 119
Fig. 7 Exploration scheme of the short term memory for the problem at hand
(B2, B3, B4) (B1, B3, B4) (B1, B2, B4) (B1, B2, B3)
this purpose, the data registered in the short memory was used. The range of the
dimensions on the bars was increased. This motivated new combinations in the
search of new neighbors, which has not been explored.
Each neighbor is evaluated with the objective function. It is verified that the
restrictions are fulfilled. Besides, it is determined if the configuration obtained
with the neighbor data is feasible. Neighbors are classified in accordance with the
evaluation of the objective function and the process selects the best candidate.
The best candidate is selected if the required movement is not prohibited from the
actual configuration to a neighbor configuration. In other words, if there is not any
taboo condition, it is feasible this sort of selection. It is called an aggressive selection.
If the best neighbor is classified as a taboo by the optimization process and it is
the best up to that moment of the analysis, the aspiration criterion lets its selection
despite its ban.
On the other hand, if there is any configuration in the neighborhood that is in
accordance with the objective function, then the best of the worst configurations
is selected. The only condition is that it had not been selected as taboo during the
optimization process. This strategy avoids the convergence at a local optimal point.
The taboo search algorithm modified the lengths of the bar. The point of initiation
is the set of initial dimensions of the bars mentioned above. Once an optimal set
was obtained, the trajectory of centrodes of the mechanism was evaluated. It was
determined the obtained error. This process continues until an optimal configura-
tion is obtained.
In the next matrix (Fig. 9), the shadowed cells show the point of initiation of
the search. From this, a random search was developed.
The taboo search started with 190 iterations and the lengths of the bars was
incremented by 0.1 mm. After the analysis of the initial results, these parameters
were changed. In the final stage, 1,900 iteration were carried on and the length of
the bars was varied 0.01 mm.
The number of iterations was incremented to improve the results in a reason-
able time. With these parameters, several alternatives were analyzed within the
dominium of the problem.
As it can be seen (Fig. 10), different combinations were obtained. In all these
cases, the trajectory of the centrode was obtained and the error was evaluated
against with the wanted trajectory of the ICR of the four bar linkage. The shad-
owed cells show the lengths of the bars of the optimized mechanism.
Optimization of the Design of a Four Bar Mechanism 121
B1 41 B2 55 B1 42 B2 57.5 B1 44 B2 57
In Fig. 11, the trajectory target of the ICR was compared against the one of the
optimized mechanism. As it can be seen, there is a good convergence. The lengths
of the links are the following: Bar1 = 43 mm; Bar 2 = 55.5 mm; Bar 3 = 29 mm
122 J. J. Muñoz-César et al.
Fig. 11 Comparison between the target trajectory of the centrode and trajectory of the ICR of
the mechanism optimized with taboo search algorithm
and Bar 4 = 59 mm. With this information a plastic model was developed with a
3D Quick prototyping system. The procedure that was followed for this purpose
is described elsewhere [13]. The prototype was evaluated in conjunction with the
orthopedic team. Afterwards, the prosthesis was manufactured (Fig. 12).
The intermediate results showed that it was possible to check different con-
figurations, as the dimensions were varied. Therefore, it was possible to observe
the performance of the algorithm. It can be considered as a good alternative with
respect to genetic algorithms, when they are used for this task [14].
The analysis was carried out in a short period. For this purpose, a personal
computer with Intel Core I3 microprocessor was used. It was required an operative
system of 64 bit and a memory of 4 GB. Besides, a great computing time was not
demanded.
124 J. J. Muñoz-César et al.
Regarding the orthopedic point of view, the individual is more confident, when
he uses the manufactured prosthesis. Initially he walked with a cane. After a cor-
rect alienation of the polycentric mechanism, the patient started to walk with both
lower limbs, in accordance with a natural gait cycle. The equilibrium was main-
tained and he has a total control over the prosthesis (Fig. 13). In accordance with
these results, it was concluded that the selected configuration of the mechanism
has a good performance and it is accordance with his anthropometric character-
istics. Security and comfort were maintained during the stance and swing phases.
Besides, the extension and flexion of the leg were adequate
7 Conclusions
This work has demonstrated the application of the taboo search algorithm in the
design of a voluntary control of a four-bar knee mechanism. In this case, several
combinations of the length of the bars were explored. All of them were within the
range of stability of the prosthesis. Convergence was observed when the obtained
trajectory of the ICR was compared with the desired one.
A random search and the transitions among de diverse solutions are the main
characteristics of this optimization process. An adequate sizing is guaranteed. This
is in accordance with the anthropometry of the individual, his physical conditions
and his pathological characteristics. The proposed methodology can be improved by
considering more relevant parameters, such as the range of forces, the dimensions
of the stump, the engagement conditions between the stump and the prosthesis,
among others. As a result a customized prosthesis can be obtained. Finally, it would
be advisable to evaluate the structural integrity of the optimized prosthesis [15].
Acknowledgments The authors kindly acknowledge the support given to the National
Polytechnic Institute, the Institute of Science and Technology of the Federal District and the
support given by the Hospital 1º de Octubre of ISSSTE.
References
1. Gunston, FH.: Polycentric knee arthroplasty. Prosthetic simulation of normal knee move-
ment. J. Bone Joint Surg. Br. 53-B, 272–277 (1971)
2. Radcliffe, C.W.: The Knud Jansen lecture: above-knee prosthetics. Prosthet. Orthot. Int. 1,
146–160 (1977)
3. Hobson, D.A., Torfason, L.E.: Optimization of four-bar knee mechanisms—A computerized
approach. J. Biomech. 7(4), 371–376 (1974)
4. Hobson, D.A., Torfason, L.E.: Computer optimization of polycentric prosthetic knee mecha-
nisms. Bull. Prosthet. Res. BRP10, 187–201 (1975)
5. Glover, F.: Tabu search–part I. ORSA J. Comput. 1(3), 190–206 (1989)
6. Glover, F.: Tabu search –part II. ORSA J. Comput. 2(1), 4–32 (1990)
7. Arora, J.S., Elwakeil, O.A., Chahande, A.I., Hsieh, C.C.: Global optimization methods for
engineering applications: a review. Struct. Optim. 9(3–4), 137–159 (1995)
Optimization of the Design of a Four Bar Mechanism 125
8. Khorshidi, M., Soheilypour. M., Peyro. M., Atai. A., Shariat Panahi, M : Optimal design of
four-bar mechanisms using a hybrid multi-objective GA with adaptive local search. Mech.
Mach. Theory 46(10), 1453–1465 (2011)
9. Smaili, A.A., Diab, N.A., Atallah, N.A.: Optimum synthesis of mechanisms using tabu-gradi-
ent search algorithm. J. Mech. Des. 127(5), 917–923 (2004)
10. Radcliffe, C.W.: Four-bar linkage knee mechanism: kinematics, alignment and prescription
criteria. Prosthet. Orthot. Int. 18, 159–173 (1994)
11. Radcliffe, C.W.: Prosthetic-knee mechanism for above knee amputees In: Murdoch, G. (ed.)
Prosthetic and Orthotic Practice, pp. 225–249. Editor Edward Arnold, LTA London, (1970)
12. Radcliffe, C.W.: Biomechanics of knee stability control with four-bar prosthetic knees. ISPO
Australia Annual Meeting, Melbourne (2003)
13. Beltrán-Fernández, J.A., Hernández-Gómez, L.H., Urriolagoitia-Calderón, G., González-
Rebatú, A., Urriolagoitia-Sosa, G., Galán-Vera, M.M., Escalante-Rodríguez, E.: Assessment
of the structural integrity of C3-C5 cervical porcine vertebrae model based on 2D clas-
sic CAD, 3D scanner and 3D computed tomography. In: Ochsner, A., da Silva, L.F.M.,
Altenbach, H. (eds.) Analysis and Design of Biological Materials and Structures. Advanced
Structured Materials, vol. 14, pp. 3–17. Springer, Berlin Heidelberg, (2012)
14. Lugo-González, E., Merchán-Cruz, E.A., Hernández-Gómez, L.H.: Synthesis optimization of
planar mechanisms. Appl. Mech. Mater. 15, 55–60 (2009)
15. Rodríguez-Martínez, R., Urriolagoitia-Sosa, G., Torres-San Miguel, C.R., Hernández-
Gómez, L.H., Urriolagoitia-Calderón, G., Carbajal-Romero, M.F.: Development of an experi-
mental apparatus for testing a total knee prostheses focused on mexican phenotype. Int. J.
Phys. Sci. 7(43), 5779–5786. (2012)
Chitosan and Poly (Vinyl Alcohol) Blends
Modified by Radiation
1 Introduction
Chitin
Fig. 1 Chitin structure
Chitosan
Fig. 2 Chitosan structure
the deacetylation of chitin and the biopolymer is regarded as the degree of dea-
cetylation (amount of deacetylated monomers in their chains) is greater than
50 %. According to the degree of deacetylation we can obtain different chitosan
samples. The properties of chitosan depend on several factors, but the molecu-
lar weight and degree of deacetylation determine its characteristics in solution
[1–6]. Is water insoluble, bases, alcohols and acetone, and completely soluble in
solutions of certain organic acids such as ethanoic acid at pH < 6. Some diluted
inorganic acids such as nitric, hydrochloric, perchloric acid or phosphoric, can
be used for preparing solutions of chitosan only after prolonged stirring and
heating [7]. The structures of chitin and chitosan are shown in Figs. 1 and 2,
respectively.
This biopolymer is insoluble in water, bases, alcohols and acetone, and com-
pletely soluble in solutions of certain organic acids such as ethanoic acid at
pH < 6. Some diluted inorganic acids such as nitric, hydrochloric, perchloric or
phosphoric acid can be used for preparing solutions of chitosan only after pro-
longed stirring and heating [8]. The solubility of chitosan is related to the amount
of protonated amine groups (−NH3+) in the polymer chain. Greater electrostatic
repulsion between the chains and also greater solvation in water [7].
PVAL is a semi-crystalline polymer obtained by polymerization of vinyl acetate
followed by hydrolysis of poly (vinyl acetate). Solubility and crystallinity influ-
ence the sensitivity to water, tensile modulus and thermal properties. The hydroxyl
groups present provides the formation of hydrogen bonds intra-and intermolecular,
causing crystallization of the polymer and thus reduces water solubility. This poly-
mer (see Fig. 3) is a material known for its chemical resistance and biomedical
Chitosan and Poly (Vinyl Alcohol) 129
Fig. 3 PVAL structure
Fig. 4 Industrial machine of
electron beam generation
properties. Since it is neither toxic nor carcinogenic, it is used for example as burn
covering, vascular prostheses, artificial membranes, and other applications in agri-
cultural and veterinary field [4, 6].
The electron beam radiation is an ionizing radiation generated by industrial
machines as show Fig. 4.
The ionizing radiation promotes the chain scission or the crosslinking in the
polymer [9, 10]. When the chain scission occurs, there is a decrease of the proper-
ties of the polymer and as consequence a decrease of molar mass average. On the
other hand, when occurs the crosslinking, there is an increase of the polymer prop-
erties and also an increase in the molar mass average [11].
A polymeric blend is a physical mixture of two or more polymers and/or copol-
ymers that was developed to increase or change the properties of each polymer
which is contained in the blend and obtain specific characteristics for different
applications [12, 13]. To be considered a blend, compounds must have a concen-
tration above 2 wt % of the second component.
The present investigation reports the formation of blends and the change in
properties, such as mechanical, thermal and water absorption after undergoing an
electron beam of radiation dose of 25 kGy.
130 M. C. Terence et al.
2 Experimental
The experimental part of this work was conducted in the laboratory of Materials
Engineering of the Mackenzie Presbyterian University (São Paulo–Brazil) and
irradiation was performed at the Institute for Energy and Nuclear Research (IPEN)
in São Paulo (Brazil).
Materials: The chitosan was purchased from the chemical company DEG,
batch 18, in the form of fine granules. PVAL was purchased by the company
Dinâmina Química, batch 24700, in the form of granules with whitish appear-
ance. The sodium hydroxide used on the deacetylation of chitosan was supplied by
Lafan química fina LTDA.
2.2 Swelling Index
The swell is defined as the amount of water absorbed by the solvent or polymeric
material when it is submerged in water for a certain period of time. The water
absorption is a function of absorption time and temperature of the solvent [14]. In
the presence of suitable solvents, polymers swell incorporating the solvent. This
process involves diffusion of solvent molecules into the polymer. Through trials
swelling is possible to determine the density of crosslinking, crosslinking effective
average molar mass between the network nodes, the coefficient of swelling and
sorption capacity [15].
The membrane was weighted (Wd) and immersed in an excess of deionized water
at ambient temperature as shown in Fig. 5. Daily the mass of the wet membrane
(Ww) was obtained after gently removing the surface water with blotting paper and
Chitosan and Poly (Vinyl Alcohol) 131
Fig. 5 Swelling tests
the process ended when no further change in weight of the material was observed.
The swelling index (SI) was then calculated on the basis of the masses of wet mem-
brane and dry membrane using Eq. (1):
(1)
SI = (Ww − Wd ) Wd × 100 %
was calculated with Eqs. (2) and (3), where the initial dry mass (Wid) is the mass
before starting the process and the final dry mass (Wfd) is the mass of the gel at
the end of 48 h, measured after the sample get dried.
(2)
Gel Fraction = (Wid − Wfd ) Wid × 100
2.4 Tensile Testing
2.5 Thermal Analysis
The dynamic mechanical thermal analysis, DMTA has been widely used in the
characterization of polymers. One of the most common uses of this technique is
the determination of glass transition temperature (Tg) and crystalline melting tem-
perature (Tm) of polymeric materials [16].
For the thermal analysis, two different tests were conducted. The first one, the
dynamic mechanical analysis (DMA) was evaluated on a Perkin Elmer DMA800,
each specimen was run from 25 °C to 160 °C at a frequency of 1 Hz and under
a tensile force. The test had an average of 45 min duration for each sample. The
second one, the differential thermal analysis (DTA) ultimate tensile strength and
percent breaking were evaluated on a Netzsch-STA409C, the specimens were
subjected to a temperature increase to analyze the Tg, and the material degrada-
tion. The test begins with an initial temperature of 25 °C and runs until 600 °C to
achieve the polymer degradation at a heating rate of 10 K/min.
3.1 Swelling Index
As can be seen in Table 1 and Fig. 7, the swelling index of PVAL and PVAL/chitosan
irradiated membranes increases up to 196 h and 220 h respectively when constant
mass is obtained.
It was noted that during the swelling test, hydrogels absorbed a huge amount
of water, so it allows us to conclude that the membrane has a great hygroscopic
Chitosan and Poly (Vinyl Alcohol) 133
character once the membrane began to disintegrate only about 10 days after the begin
of the analysis. It was also possible to analyze that the membrane that contains only
PVAL, without chitosan, showed better results in terms of ability to absorb water.
Normally, the grown crystalline portion in a chitosan membrane prevents water from
entering membrane and the larger the crystallinity of a chitosan membrane, the
smaller the swelling index of the membrane [17].
The results obtained for the gel fraction and the sol fraction in the polymeric blend
of chitosan and PVAL and in the neat PVAL membrane are showm in Fig. 8. All
samples were irradiated.
134 M. C. Terence et al.
3.3 Tensile Testing
Table 2 shows the results obtained from the tensile test. The membrane irradiated
has a greater deformation to chitosan/PVAL.
During the test and with the results it was verified that crosslinking occurred.
The samples irradiated presented higher final deformation.
Figure 9 shows the results obtained during the DTA tests with the different speci-
mens that were used as the irradiated PVAL membrane, the chitosan and PVAL
irradiated membrane and the chitosan and PVAL membrane non irradiated.
Analyzing the graph it is possible to see that once again the amount of
crosslinking interfere with the degradation of the material as well as in his behav-
ior with an increase of the temperature. It is possible to observe that the specimen
with higher crosslinking in the structure shows higher Tg and also that crosslinking
influences directly in the degradation of the material. Also it is possible to verify
that the chitosan increase the Tg of the material.
Chitosan has contributed to the lifetime of the hydrogel resulting in a greater
resistance and stability with an increase of the temperature.
Tan Delta
Figure 10 shows the results obtained with the PVAL irradiated membrane and
Fig. 11 shows the results obtained with the chitosan and PVAL irradiated membrane.
The crosslinking material shows great resistance with temperature variations
and better elastic properties.
136 M. C. Terence et al.
Modulo /Pa
2995ral 0.400
Tan Delta
2721ral 0.300
2447ral 0.200
2173ral 0.100
1900ral 0.000
00 50 100 150 200
Temperatura /°C
4 Conclusion
With the results obtained in this work, that analyzes the effect of the radiation in
a polymeric blend, it is possible to say that with the method adopted and with the
dose of radiation used, crosslinking occurs.
According to the tensile tests, to analyze the mechanical properties of the
PVAL, it was possible to verify that has excellent properties and that when sub-
jected to electron beam radiation acquires greater elasticity becoming more mal-
leable. It was also observed that during the test occurs the crystallization of the
material due to the orientation of the chains and looking directly to the specimen
after rupture the cross section became whitish.
With the swelling index was possible to verify that the specimens that were not
subjected to radiation have been dissolved completely in less than 24 h due to the
absence of crosslinking.
The thermal analysis has proven that the Tg changes due to the miscibility of
the materials and that the crosslinking increases the Tg. Also was observed that
when the material is subjected to mechanical deformation changes in Tg, Tm and
in the temperature of material degradation is noticed.
Radiation therefore interferes directly with the capacity of water absorption and
deformation under tensile.
References
1. Rinaudo, M.: Chitin and chitosan: properties and applications. Prog. Polym. Sci. 31, 603–632
(2006)
2. Goosen, M.F.A.: Applications of Chitin and Chitosan. Technomic, Basel (1997)
3. Terence, M.C., Miranda, L.F., Faldini, S.B., Castro, P.J.: Study of chitosan addition in the
PVP/PVAL polymeric blend. Adv. Sci. Tech. 76, 190–194 (2010)
4. Rodrigues, I.R.: Síntese e caracterização de redes poliméricas a base de quitosana com PVP
e PVA para aplicação na liberação controlada de fármacos. Universidade Federal do Rio
Grande do Sul, Rio Grande do Sul, Dissertação de mestrado (2006)
5. Terence, M.C., Faldini, S.B., Miranda, L.F., Munhoz A.H., Jr.,, Castro, P.J.: Preparation
and characterization of a polymeric blend of PVP/PVAL for use in drug delivery system. J.
Biomed. Nanotech. 7, 446–449 (2011)
Chitosan and Poly (Vinyl Alcohol) 137
6. Peppas, N.A., Bures, P., Leobandung, W., Ichikawa, H.: Hydrogels in pharmaceutical formu-
lations. Eur. J. Pharm. Biopharm. 50, 27–46 (2000)
7. Barros, F.C.F., Carvalho, T.V., Cavalcante, R.M., Dias, F.S., Nascimento, R.F., Queiroz, D.C.,
Vasconcellos, L.C.G.: Produção e caracterização de esfera de quitosana modificada quimica-
mente. Rev Iberoamer Polím 7, 232–246 (2006)
8. Santos, J.E., Soares, J.P., Dockal, E.R., et al.: Caracterização de quitosanas comerciais de
diferentes origens de polímeros. Polímeros: Ciência e Tecnologia 13: 242–249 (2003)
9. Terence, M.C.: Obtenção de um sistema de liberação controlada de drogas a partir do PVAL
irradiado com radiação gama. Tese de Doutorado, IPEN, São Paulo (2002)
10. Rosa, D.S.: Modificação de polietileno de baixa densidade, polipropileno isotático e
suas blendas, por radiação gama. Universidade Estadual de Campinas, Campinas, Tese de
Mestrado (1991)
11. Wan Y, Creber KAM, Peppley B, Tam Bui V (2003) Synthesis, characterization and ionic
conductive properties of phosphorylated chitosan membranes. Macromol Chem Phys 204:
850–858
12. Callister W.D. Jr.,: Material Science and Engineering: An Introduction. Wiley, New York
(2004)
13. Felisberti, M.I.: Apostila de caracterização de blendas poliméricas através de análise térmica
e termomecânica. Universidade Estadual de Campinas, Campinas (1998)
14. Miranda, L.F.: Estudo de parâmetros de processo para a síntese de membranas hidrofílicas a
base de poli(N-vinil-2-pirrolidona. Tese de Doutorado, IPEN, São Paulo (1999)
15. José, N.M., Souza, C.M.L., Felisberti, M.I., Yoshida, I.V.P.: Síntese de redes a base de
PDMS e reticulantes dendríticos. Anais do Congresso Brasileiro de Engenharia e Ciência dos
Materiais, São Pedro (2000)
16. Canevarolo S.V. Jr.,: Análise térmica dinâmico-mecânica. Canevarolo SV Jr Técnicas de car-
acterização de polímeros, Artliber, São Carlos (2004)
17. Oliveira, MJA., Rodrigues, KMS., Oliveira, TF., Parra, DF., Lugão, AB.: Hidrogéis de
poli(álcool vinílico)/poli(n-2-pirrolidona) para liberação de fármaco. In: 9 Congresso
Brasileiro de Polímeros, Campina Grande (2007)
Mutual Connections Between Mechanical
and Material Factors, and the Biological
Processes of Implants Adaptation
Abstract The paper presents problems connected with the influence of mechanical
and material factors on the biological adaptation processes of femoral bone implants.
A frequently used type of hip joint implant was chosen as the example. These mod-
els were loaded in particular stages of the human gate and then they were used
for the analysis of stress changes. The identification of the relations between the
mechanical properties of osseous tissue required the conducting of computer simula-
tions by means of the finite element method (FEM).
1 Introduction
A. Jasik (*) · M. Jabłońska
Faculty of Materials Engineering and Metallurgy, Silesian University of Technology,
Krasinskiego 8, 40-019 Katowice, Poland
e-mail: anna.jasik@polsl.pl
M. jabłońska
e-Mail: magdalena.jablonska@polsl.pl
2 Method of Research
For the stress analysis in an artificial hip joint, a non-cement Fitmore system from
Zimmer [5] was chosen. The Fitmore endoprosthesis was modeled and mounted
in the bone using the Alibre Design software. It is an innovative CAD system for
solid modeling and intended for designing both complex sets of elements and their
individual parts [6].
Alibre Design cooperates with the Algor software, enabling creations of a grid
and models of finite elements using tools and creators for enhancement of effi-
ciency and accuracy of the simulation, and carrying out a stress pattern analysis.
The Algor software merges CAD designing with FEM analysis, and it enables
the analysis of models’ performance under temperature or load with selection of
proper boundary conditions [7].
In the Alibre Design software, three types of Fitmore endoprostheses [5] were
modeled:
• A with neck inclination angle 140º,
• B with neck inclination angle 137º,
• C with neck inclination angle 127º.
Geometrical features of the models were selected based on the Zimmer
company catalogue.
In Fig. 1, femur models after implantation of Fitmore stems A, B, and C are shown.
In femur models, various properties of femur-implant system components
shown in Table 1 were assumed.
Mutual Connections Between Mechanical and Material Factors 141
Fig. 1 Femur model after implantation (a) coronal plane view, (b) sagittal plane cross-section
The models were fixed and loaded. For the endoprosthesis head, contact area
conforming with the anatomical one was assumed and the head was loaded with
pressure with value corresponding to natural load of femur head. In the designed
endoprosthesis, influence of forces of abductor muscles Qm and Pm with various
directions and values acting onto greater trochanter of the femur were taken into
account. In Table 2 [8], load values of a natural femur model with implanted endo-
prosthesis for 16th phase of human gait are shown.
142 A. Jasik and M. Jabłońska
Numerical calculations carried out using the Algor software enabled to determine
stress patterns for three types of Fitmore endoprostheses.
The stress patterns were determined according to the Huber hypothesis (Eq. 1),
used for materials with various properties—both plastic and brittle materials [9].
1 2 2
σred = √ σx − σy + σy − σz + (σz − σx )2 + 6 τxy 2 + τ2 + τ2
yz zx
(1)
2
The implants are intended for providing proper conditions for reconstruction of
damaged osseous tissue or complete the adoption of its function. Analyzing the
stress patterns (Figs. 2, 3) for an artificial hip joint, one may ascertain that the
enroprosthesis stem transfers the loads, adopting partially the function of the bone.
The highest stress concentration in endoprostheses occurs in the lower part of
the stem, in the neck, and in the area of Adams arch (thickened medial part of
the femoral neck, being a support for the prosthesis). One may conclude that the
higher the inclination angle of the prosthesis neck, the higher stress concentration
in this area.
Numerical analysis proved that depending on the inclination angle and stem
neck length of the Fitmore prosthesis, the stress pattern changes in the area of the
so-called Adams arch. It was noted that in the case of stems with an inclination
angle of 127º, the stresses assume lower values in comparison to stems with neck
inclination of 137º and 140º.
For stems with an inclination angle of 140º, the prosthesis neck, where reduced
stresses are concentrated, is the most stressed element.
For an implant with an neck inclination angle of 137º, the stress pattern in its
area are intermediate while compared with the other analyzed models.
In the case of Fitmore A stem, the highest stress concentration is located on its
top. The stem of the Fitmore B prosthesis in maximally loaded along its lateral
lower edge. While the stem of the Fitmore C prosthesis is maximally loaded along
the Adams arch.
Stress patterns after implantation of all types of Fitmore stems in femur are
evenly. Stress concentration in the bone at the base of the model is caused by its
fixing in the Algor software.
For a detailed presentation of stress values, areas designated by numbers 1 and
2 were selected for models of endoprosthesis stems and natural bone after implan-
tation (Figs. 4, 5). The areas describe the effect of an endoprosthesis stem on osse-
ous tissue.
The highest values of average reduced stresses occurs in the Fitmore A prosthe-
sis both in area 1 (Fig. 6a) and area 2 (Fig. 6b). Therefore, it may be ascertained
that the majority of stresses is transmitted by the prosthesis. In the case of models
144 A. Jasik and M. Jabłońska
Fig. 4 Selection of area No. 1 for detailed stress analysis (marked with pink dots): (a) in
Fitmore C endoprosthesis, (b) in the bone. Coronal plane view
Fig. 5 Selection of area No. 2 (marked with pink dots) for analysis of values of mean reduced
stresses σz: (a) in Fitmore A endoprosthesis, (b) in the bone. Coronal plane view
C and B, in the upper part of endoprostheses (area 1) higher reduced stresses occur
than in the bone, and they are almost twice as high.
Mutual Connections Between Mechanical and Material Factors 145
Fig. 6 Cumulative list of mean reduced stress σz for endoprostheses and bones in area: (a) 1; (b) 2
In the case of the bone, higher reduced stresses occur in the lower part (area 2)
than in the upper one (area 1). Thus the bone is unloaded in the upper part, and
loaded in the lower part. It is the most undesirable effect.
Implantation of an endoprosthesis that leads to unloading of the trochanters
area and to transmission of loads by osseous tissue around the lower section of
the stem, may contribute to osseous decay, called stress shielding (decrease in
loads inside osseous tissue). Osseous resorption may be a symptom of low level of
reduced stresses in the upper part of the bone (according to assumptions presented
by Wolf—“trabecular structure of osseous tissue under equilibrium conditions
adapts to directions of principal stresses”) [10]. It means that change in load and
change in action of force vector directions results in adaptation of bones to current
state of stresses. This phenomenon may be connected with the fact that modern
endoprostheses are more rigid than cortical bone and they deprive some areas of
osseous tissue of proper stresses being a stimulator of its adaptation do variable
environment. That is why introduction of an implant often causes a takeover of all
transmitted loads, what may mean an “undernutrition” of the bone and its faster
dissolution (ostheolysis), possibly leading to a fracture.
While analyzing area 2 for Fitmore B prosthesis (77.33 MPa) and for the bone
(81 MPa) it was noticed that stresses reach comparable values. It is the most desir-
able state, as it is pursued that the prosthesis and the bone transmit similar stresses,
in order to avoid the bone decay and destruction of the prosthesis.
4 Summary
References
1. Jasik, A., Okrajni, J.: Attempt to evaluate the influence of the thickness of a surgical cement
layer on adaptation processes of endoprosthesis stems. Problems of modern techniques in
aspect of engineering and education. Monography. pp. 261–266. Pedagogical University
Cracow (2006)
2. Jasik, A., Ziemba, S., Plaza, M.: Methodology of estimation of the hest flow during endo-
prosthesoplasty. Eng. Biomater. 10(69–72), 99–101 (2007)
3. Okrajni, J., Jasik, A., Kusz, D. et al.: The influence of material property nonhomogenity and
mode of loading on the effort of femur model components. Acta Bioeng. Biomech. 3(Suppl. 2),
379–386. Wroclaw University of Technology Publisher, Wrocław (2001)
4. Han, S.M., Chu, J.U., Park, S.H., et al.: Finite element analysis of mechanical stability of
ceramic acetabular components and evaluation of rom in articulating hip joints. J. Biomech.
Sci. Eng. 6(3), 173–182 (2011)
5. Brochure Fitmore Hip Stem. USA (2009)
6. www.alibre.pl [22. 04. 2011]
7. www.algor.info.pl [22. 04. 2011]
8. Bombelli, R.: Ostheoattists of the Hip. Classification and Pathogenesis the Role of
Osteotomy as a Consecuent Therapy. Springer-Verlag, Heilderberg (1983)
9. Jakubowicz, A., Orłoś, Z.: Strength of Materials. Science-Technical Publisher, Warszawa
(1968)
10. Kusz, D.: Biomechanical aspects of hip endoprosthroplasty. Inżynieria Materiałowa. 2, 39–44
(1998)