HUMAN KNEE PROSTHESIS EQUIPPED WITH FORCE
SENSORS
M. Gazzoli, E. Sardini, M. Serpelloni
Department of Information Engineering, University of Brescia, Brescia, Italy
G. Donzella
Department of Mechanical Engineering, University of Brescia, Brescia, Italy
Keywords: Measurement in human knee, Instrumented implant, Biomechanics, Force transducer, Strain sensor.
Abstract: In-vivo monitoring of human knee implants after total arthroplasty increases the knowledge about articular
motion and loading conditions. The proposed knee prosthesis equipped with force sensors executes force
measurements in polyethylene human knee prosthesis by two sensorized metal bars positioned under the
femoral condyles and fully contained in the polyethylene insert. In this paper, a new force sensor, which
acts in the knee integrated in the prosthesis, is proposed with the aim of reducing the hysteresis of
polyethylene material and increase the rigidity of the insert. The fabricated sensors are characterized and
tested by means of a mechanical press controlled load. The realized conditioning electronics is done by low
power components and it can be integrated in an autonomous system. The forces transmitted across the knee
joint during normal human activities such as walking, running or climbing can be directly measured.
Furthermore, the device can be used to improve design, refine surgical instrumentation, guide post-operative
physical therapy and detect human activities that can overload the system.
1 INTRODUCTION
The measurement of forces acting on knees, while
walking or during normal movements of the leg is a
topic widely discussed in (Gattiker, Umbrecht,
Neuenschwander, Sennhauser, Hierold, 2008),
(Westerhoff, Graichen, Bender, Rohlmann,
Bergmann, 2009), (Heinlein, Graichen, Bender,
Rohlmann, Bergmann, 2007). In Heinlein et al.
(2007), a measurement system detects the forces
using strain gauge sensors: the authors propose the
measurement of six degrees of freedom of the knee
(three forces and three moments) using a titanium
frame inserted in the tibia with six strain gauges. In
(Crescini, Sardini, Serpelloni, 2009), an
instrumented knee prosthesis is proposed for
measure the forces between the femur and tibia by
magnetoresistors. These sensors measure the
deformation of the polyethylene surface of the
prosthesis in contact with the condyles of the tibia.
But, in this configuration and particular design, the
material shows hysteresis after application of forces,
because the polyethylene material has a viscous-
elastic component. The proposed system relies on
the use of a new sensors technology. A typical
prosthesis for human knee is composed by three
principal inserts. In Figure 1, the 3D model of the
adopted prosthesis is represented. The femoral and
tibial insert are made on titanium and the tibial plate
is made of polyethylene. On the top of the implant
two excavations are carried out (Figure 1) centered
under the contact point between the condyles and the
polyethylene insert. These holes are made for
accommodate two sensorized metal bars. Two strain
gauges are placed under the metal bars for
measuring the deformations and correlate the
measured data to the force applied. In this way, the
sensors are closer to the focal point of the condyles
with the prosthesis, and this placing allows
integrating electronics and sensors in the
polyethylene insert. The realized sensors and
conditioning electronics is done by low power
components and it can be integrated in an
autonomous system as proposed in (Crescini et al.,
2009).
349
Gazzoli M., Sardini E., Serpelloni M. and Donzella G..
HUMAN KNEE PROSTHESIS EQUIPPED WITH FORCE SENSORS.
DOI: 10.5220/0003154603490352
In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2011), pages 349-352
ISBN: 978-989-8425-37-9
Copyright
c
2011 SCITEPRESS (Science and Technology Publications, Lda.)
Figure 1: Section of upper polyethylene insert with
integrated force sensors.
2 SENSOR
With the aim of reducing the hysteresis of
polyethylene material, in this paper, a new force
sensor that acts in the knee integrated in the
prosthesis is proposed.
The polyethylene insert was
divided in two parts: top and bottom. On the upper
insert two excavations are made, one in the medial
side of the tibial plate and the other in the lateral
side. Two sensorized metal bars are placed on the
hole as shown in Figure 1. The excavations are made
exactly under the contacts point of condyles for
directly transmit the forces on the sensors. In the
bottom insert the conditioning electronic is placed.
The insert was processed with a numeric control
milling machine. The excavations have dimensions
20 mm x 20 mm. The metal bars are made by inox
steel AISI 430 with thickness of 0.8 mm with the
same dimensions of the holes. The two metal bars
are sensorized by strain gauges (with resistance of
120 , Gauge factor 1.88, dimension 5 mm x 5 mm
commercialized by HBM), and placed under the bars
as shown in Figure 1. With the aim of transferring
directly the forces on metal bars, the femoral
condyles are placed directly over the metal bars. To
ensure the biocompatibility of polyethylene insert,
on the superior face of the insert 2 mm thick film of
polyethylene was placed for covering the metal part
of the fabricated sensors. When a force is applied
through the femoral condyles, the metal bars are
bent. Young's modulus of inox steel AISI 430 is
196000 N/mm
2
and the Young’s modulus of
polyethylene is 2600 N/mm
2
. Increasing the
Young’s modulus of bent material, the material’s
hysteresis has been reduced.
Another advantage by this configuration is to
increase the rigidity of the insert. The deformation of
metal bars is very small, so that the mechanical
properties of the insert are very similar to the not
sensorized insert.
The resistances of strain gauges are measured by
inserting the strain gauges in two Wheatstone
bridges. The measured differential voltage of each
bridge was amplified and sampled by a digital
multimeter connected by GPIB bus with a PC. The
differential voltage is proportional to the force
applied by the femoral condyles.
Figure 2: Conditioning electronics of integrated sensors.
In Figure 2, the conditioning electronics of
integrated sensors are show. The systems are
composed by two Wheatstone bridges. The R
s
variable resistors are the strain gauge applied to the
metal bars, while the R resistors are the fixed
resistors of Wheatstone bridge. R
g
resistors are used
to fix the gain of instrumentation amplifiers. The
differential voltages are measured and amplified by
a micro power dual instrumentation amplifier,
INA2126, produced by Burr-Brown. The INA2126
has two precision instrumentation amplifiers for
accurate, low noise differential signal acquisition.
The two-op-amp design provides excellent
performance with very low quiescent current.
The output voltages are measured by two digital
multimeters (Fluke HP8810).
LateralHole
MetalBars
MedialHole
FemoralInsert
TibialInsert
StrainGauges
LOAD
Polyethylene
LOAD
BIODEVICES 2011 - International Conference on Biomedical Electronics and Devices
350
3 EXPERIMENTAL SETUP
The sensorized polyethylene insert is tested and
characterized by an Instron 8501 force machine. The
polyethylene insert was placed on the fixed part of
the Instron machine and the femoral component was
placed on the mobile component and the forces were
applied. The experimental setup is shown in Figure
3.
Figure 3: Experimental setup with Instron machine and
measurement system.
The force control is made by a load cell integrated
and controlled by the Instron machine. The Instron
machine has an output voltage proportional to the
applied force. This voltage is measured by the digital
multimeter (Fluke HP8810). It lets you know the
profile of force value applied to the polyethylene
insert.
Figure 4: Block diagram of the experimental setup.
In Figure 4, the block diagram of experimental setup
adopted to test the realized sensors was represented.
The two strain gauges are placed in a Wheatstone
bridge, the differential voltages are measured by two
instrumentation amplifiers and the measured
voltages are sampled by two digital multimeters
(Fluke HP8810) connected by GPIB bus to PC. The
Instron machine has an output voltage proportional
to the generated force. The voltage is sampled by a
digital multimeter (Fluke 8810), connected by GPIB
bus to PC. Dedicated software LabVIEW is realized,
it allows to acquire the data and store the differential
voltage (Medial and Lateral Sensors) and the voltage
proportional to the force applied in single file. The
Instron machine is programmed to apply to the
polyethylene insert a very slow variation of load
forces. The applied forces start from 300 N to 1000
N. This force profile is used for the sensor
characterization. Using the experimental setup
showed in Figure 4 the Medial and Lateral Sensor
are characterized. The Instron machine is
programmed to generate the force profile previously
described. The measurement information are stored
in a single file and digitally elaborated.
The data obtained are showed in Figure 5. For
clarity of content only the characterization of Medial
Sensor was reported. The results of Lateral Sensor
are the same.
The output voltage from instrumentation amplifiers
is proportional to the applied force, and the
relationship between force and voltage is linear.
Figure 5: Medial sensor characterization.
The measured voltage was interpolated with a linear
straight and his equation (1) that correlates the
voltage with applied force is:
F = 593392·V-179064 (1)
Where F is the force applied in newtons (N) and V is
the measured voltage in volts (V).
4 CONCLUSIONS
In this paper the authors propose a different sensor
technology for force measurement in a total
prosthesis of human knee and it is fully contained in
the polyethylene insert. The proposed sensors
consist in two metal bars positioned under the
femoral condyle. The metal bars are made on inox
steel AISI 430 because it has a high Young’s
GPIB
Fl
u
k
e
8810
Fluke8810
Fluke8810
LabVIEW
Instron
8510
Medial
Sensor
LabView
Multimeter
Amplifier
Instron
Machine
GPIBBu
s
Multimeter
Lateral
Sensor
Multimete
r
Amplifier
HUMAN KNEE PROSTHESIS EQUIPPED WITH FORCE SENSORS
351
modulus. That allows reducing the material’s
hysteresis. The metal bars are sensorized by two
strain gauges. The sensor characterization s was
done and the obtained data show a linear
relationship between the applied forces and the
measured voltages.
These sensors can be integrated in an
autonomous system that does not require internal
power sources, such as batteries. The autonomous
system can integrate an energy harvesting module
that extracts energy from magnetic field.
The forces transmitted across the knee joint
during normal human activities such as walking,
running or climbing can be directly measured.
Furthermore, the device can be used to improve
design, refine surgical instrumentation, guide post-
operative physical therapy and detect human
activities that can overload the system.
The sensorized total knee prosthesis will be
tested by a robotic mannequin for dynamical
characterization. The robotic mannequin can
simulate the human walking. During the test the
polyethylene insert will be subjected to dynamic
forces and it will be possible to make a pseudo in-
vivo dynamic characterization of the sensors.
REFERENCES
Gattiker, F., Umbrecht, F., Neuenschwander, J.,
Sennhauser, U., Hierold, C., 2008. Novel ultrasound
read-out for a wireless implantable passive strain
sensor (WIPSS), Sensors and Actuators A 145–146
291–298.
Westerhoff, P., Graichen, F., Bender, A., Rohlmann, A.,
Bergmann, G., 2009. An instrumented implant for in
vivo measurement of contact forces and contact
moments in the shoulder joint, Medical Engineering &
Physics 31 207–213.
Heinlein, B., Graichen, F., Bender, A., Rohlmann, A.,
Bergmann, G., 2007. Design, calibration and pre-
clinical testing of an instrumented tibial tray, Journal
of Biomechanics 40 S4–S10.
Crescini, D., Sardini, E., Serpelloni, M., 2009. An
Autonomous Sensor for Force Measurements in
Human Knee Implants, Proceedings of the
Eurosensors XXIII conference Procedia Chemistry
1(1) 718-721.
BIODEVICES 2011 - International Conference on Biomedical Electronics and Devices
352