A platform based foot pressure/shear sensor

Chun-Te Changf, Chao shih Liue ,William Soetantoc , Wei-Chih Wanga,b,d*

a
Department of Mechanical Engineering, University of Washington, Seattle, WA
b
Department of Electrical Engineering, University of Washington, Seattle, WA
c
Information School, University of Washington, Seattle, WA
d
Medical Device Innovation Center, National Cheng Kung University, Tainan, Taiwan
e
Department of Power Vehicle and Systems Engineering, Chung Cheng Institute of
Technology, National Defense University, Taoyuan, Taiwan
f
Department of Electronic Engineering, Southern Taiwan University, Tainan, Taiwan
.

ABSTRACT

The proposed research is aimed at developing, fabricating and implementing a flexible fiber optic bend loss sensor for the
measurement of plantar pressure and shear stress for diabetic patients. The successful development of the sensor will
greatly impact the study of diabetic foot ulcers by allowing clinicians to measure a parameter (namely, shear stress) that
has been implicated in ulceration, but heretofore, has not been routinely quantified on high risk patients. A full-scale foot
pressure/shear sensor involves a tactile sensor array using intersecting optical waveguides is presented. The basic
configuration of the optical sensor systems incorporates a mesh that is comprised of two sets of parallel optical waveguide
planes; the planes are configured so the parallel rows of waveguides of the top and bottom planes are perpendicular to
each other. The planes are sandwiched together creating one sensing sheet. Two-dimensional information is determined
by measuring the loss of light from each of the waveguide to map the overall pressure distribution. The shifting of the
layers relative to each other allows determination of the shear stress in the plane of the sensor. This paper presents latest
development and improvement in the sensors design. Fabrication and results from the latest tests will be described.
.
Keywords: plantar pressure/shear sensor, shear sensor, fiberoptic sensor, bend loss sensor, stagger-structure sensor.

1. INTRODUCTION
Shear force is an important parameter to evaluate how ulcer is developed for diabetic patient. Most commonly reported
techniques utilize electromagnetic or piezoelectric means to deduce this parameter. Previously, attempts have been made
using capacitive sensors [1,2,8,9]. The ideas behind capacitor sensors are similar. They measure the horizontal
displacement based on the relative capacitance change due to the change in the area of overlap between the top and
bottom electrodes. To improve the shear compliance and flexibility, all three reported methods used
polydimethylsiloxane (PDMS) were as mechanical support and substrate. The advantages of these techniques are they
are relatively easy to understand and implement. However, these sensors made of polymers are lack in a robust
manufacturing process and efficient multi-channel capacitance measuring circuit. Currently, these devices all require
some manual assembly in construction, which makes future mass production very difficult. To reduce the size of the
sensor, a low noise high precision multi-channel detecting electronics is also needed. For a 0.1x1x1mm3 sensor, the
electronics will require a 0.1pF or better in sensitivity. It is extremely difficult to achieve this resolution without using an
IC based electronics such as capacitor-to-digital converter. Currently there isn't one available for more 2 channels.
Using these techniques also require a very fast DAQ card and smart multiplexing techniques. For the resistive and
piezoelectric based shear sensors, these sensors suffer from the same problems as the capacitive sensors in lack of proper
electronics support. It also suffers from thermal drift, electrical and mechanical hysteresis, and electromagnetic
interference problems that plague most electromagnetic sensors, which we have discussed in our previous papers [3, 4,
5].
To prevent electromagnetic interference caused by the surroundings and also by the human body, Missinne [10], has
reported an optical tactile sensors similar to the current capacitive sensors design, where shear and normal displacement

Health Monitoring of Structural and Biological Systems 2012, edited by Tribikram Kundu, Proc. of SPIE Vol. 8348,
83481U · © 2012 SPIE · CCC code: 0277-786X/12/$18 · doi: 10.1117/12.915396

Proc. of SPIE Vol. 8348 83481U-1

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 10/15/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

 is observed by the relative change in the displacement between the top and bottom sensing layer. Herein, displacement in
x, y and z directions are detected by an optical pickup made of vertical-cavity surface-emitting lasers (VCSEL) as the
light source and photodiodes as the detectors. This design however still required a large number of output channels for
distributive sensing.

Previously, we have reported a distributive pressure/shear sensor using bend loss sensor [3, 4, 5, 6, 7, 11, 12]. The sensor
is composed of two optic fiber meshes which are embedded into a PDMS pad so that fibers’ radius of curvatures change
when the sensor is pressed. The measurement of light intensity change through bending each fiber provides information
about the fiber’s changes in curvature. Shear is measured based on the relative position change in the pressure point
between the top and bottom layers of the sensor. This sensor has several advantages. One is that it can provide the same
number of discrete pressure points compared to commercially available capacitive or resistive based pressure sensors
while this optical sensor requires only 2 times the square root of the total number of sensors as the capacitive or resistive
sensors. This is due to the fact that each pressure point is defined by an intersection of a row and a column optical
waveguide. Therefore the number of output channels is reduced to the total number of waveguides used for the sensor.
This unique feature reduces the data acquisition time for each output channel.

In this paper, a 30 channels (20 x 28 cm2) plate form based sensor is presented. Some measurement of the shear forces
on the foot heel of several different condition feet are discussed. In this paper, we will present our preliminary results
from the shear and pressure measurement of an unhealthy hammer toe foot .
.

2. METHODOLOGY

The sensor is made of two layers of fiber mesh that are embedded in PDMS polymer. The pressure and shear are
measured based on intensity modulation in the fibers when the PDMS is compressed. When the fiber is bent, it
introduces bend loss, which leads to a decrease in the light intensity passing through the fiber. In addition, the fiber
staggered design, where two fiber meshes are stacked but the positions of the top grid layer and bottom grid layer are
offset, makes the shear force measurement possible. These two layers work in conjunction to be able to identify both
normal and shear forces. When we apply a normal force on the top layer of the sensor, the fiber would displace straight
down, which will not cause the fiber mesh underneath it to bend. On the other hand, when we apply a shear force the top
fiber will displace along the direction of the force applied. This will cause the fiber along that direction to bend and
create the intensity loss along the top and bottom layer (Figure 1).

Sensing area

Foam 1537

(a) (b)

Figure 1. (a) Fiber meshes layer out inside the full size foot sensor (b) Side view of the sensor showing the fibers being
put in stagger position.

3. FABRICATION
The supporting frame for the sensor is made of 1.5 mm thick acrylic board cut and etched by a CO2 laser with small
narrow trenches of 200um deep and wide and spaced 5 mm apart for fiber coupling grooves. Large 200um core
multimode fibers (Fiberguide Industries AFS 200/220N) are inserted and secured into the trenches. The spacing between
fibers in the sensor area is 2 cm apart. The sensor is made of a 7x7 mesh for the top layer and an 8x8 mesh for the

Proc. of SPIE Vol. 8348 83481U-2

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 10/15/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

 bottom layer. The mesh is encapsulated in a flexible polymer (Smooth-On Ecoflex 0030) creating a 3 mm thick sensing
layer. To protect the sensing layer, we then pour one more 1.5 mm thick layer of Ecoflex 0030, and also a 1.5 mm thick
layer of Smooth-On Oomoo 30. After all of the layers are cured, the sensor is placed on a foam mat (EARSC 1537),
which serves as a cushion for the sensor to compress and allowing the fiber to bend further. LED/light source and photo
detectors are attached to the two ends of each fiber to power the fiber and measure the light transmission. Infrared light
emitting diodes (LEDs) are used, which produce 940nm IR light (TSAL 4400) to eliminate interferences from visible
light during data collection. The light intensity as measured by the intensity to voltage converter (TSL 260RD) is then
collected (using NI USBDAQ 6259) and processed using National Instruments Labview software. The measurement of
light transmission through the fiber intersection points would provide information about magnitude of the force and the
bend loss characteristic of the fiber. The structure of the sensor is shown in Figure 2.

Photo Detectors
In shield boxes
Roll up Oommoo
protecting layer
PDMS
Elastomer covered fiber
mesh sensor area
Optic fibers

Acrylate frame with

Optic fibers
fiber grooves

Potentiaometer

LEDs

Figure 2. The structure of the sensor: An assembled shear sensor along with its electronic components

In previous paper [7], the shear sensor detector assembly is constructed using the TSL260R-LF photo detector. While
this particular detector can do the job there are certain drawbacks. One such drawbacks is the lens of the detector itself.
As the fiber hit the lens of the photo detector lens the fiber have a tendency to deflect away from the lens causing the
fiber to shoot the light signal away from the lens and away from the detector. One possible solution to this problem is to
integrate a design from the previous detector coupler design. We can use the wire insulator and heat shrink to create a
channel which will allow the fiber to couple head on with the detector lens. With the sleeve it would be impossible for
the fiber to be deflected from the lens of the detector and should improve coupling. While this might work in the short
term it would be better to change the design completely to use a different photo detector completely that will suit our
design. We have found a suitable replacement for said part which is the TSL260RD. TSL260RD photo detector has the
same underlying sensor as the TSL 260R-LF except it is packaged in a SOIC-8 SMD package. This package does not
have a lens instead it favors the use of a window with a diameter of 2.8mm. As it uses a window it would be easier for
the fiber to couple the light to the sensor without having very precise coupling. Another advantage of using the
TSL260RD is consistency that we can achieve. As the SOIC-8 packaging is an SMD package it means we do not have to
cut any leads that can change the dimensionality of the device. With the PCB pads exposed there will only be one way to
solder the device and the device should be aligned in a very consistent manner. With this consistency we can use the
laser machine to cut the pieces to ensure a good fit overall with the sensor itself. One such design for the detector
modules is shown in Figure 3. The design itself requires several different parts a patterned PCB board, the TSL260RD
sensor itself, 4 copper rods 22AWG, and an interface between the sensor and the fiber itself.

Proc. of SPIE Vol. 8348 83481U-3

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 10/15/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

 Figure 3. Left is a fully assembled view of the detector and right is a exploded view of the detector module

4. EXPERIMENTAL METHOD
Previously [7], we used small probes to evaluate the mechanical and optical performance of the newly
designed sensor. The result shows the sensor is able to detect both shear and pressure stresses.
Ultimately, this sensor is to allow quick identification of different foot patterns especially near the heel area
and its corresponding stress distribution. Tests were conducted on prosthetic feet with three different foot
soles (Figure 4). Silicone and Gelatin were used as the main materials for the feet construction except for the
flat foot which is a prosthetic foot cover. The general dimension of these three feet is shown in Figure 5.
Different foot shape will create different stress distribution when it is on the sensor. For example, a flat foot
will have the largest contact area with than a normal healthy foot. The special foot in Figure 4(c) has
smallest contact area at the ball and heel area for simulating diabetic patients’ ulcer foot or people with
hammer toe.

(a) (b) (c)

Figure 4. Different kind of full size prosthetic foot (a) Normal healthy foot, (b) Flat foot, (c). Hammer toe foot.

Proc. of SPIE Vol. 8348 83481U-4

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 10/15/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

 Figure 5. The general dimension of prosthetic foot

The full scale foot is then connected to an Interface 1210 AJ-2 load cell (2500N max load, .5 N resolutions) controlled
by Instron 5585H load frame for testing. The applied force is calculated using equation (1) by assuming a person with
100kg weight is standing on one foot. Using equation (1), we are able to obtain the necessary force of 132 N for special
foot, 151 N for flat foot and 187 N for normal foot as tabulated in Table 1. We include the constant of 1.1 to take into
account of the ground reaction force which acts upon the sensor as a person steps or walk upon the sensor.

. (1)

Table.1 Generalization for the full foot force

Foot shape Foot / Heel Area (cm2) F(N)
Normal foot (man size 10) 115/ 20 187
Full Flat foot 178/25 151
Special foot 65/8 132
The overall dimension of the foot sensor is 20 by 30 cm2. There are total of 113 pressure points between these two
sensing layers. Two fiber meshes are put perpendicular to each other and the gird layout of top and bottom layers is
offset as shown in Figure 6. The top layer has 7 by 7 fibers to create 49 crossing points and the bottom layer has 8 by 8
fibers to create 64 crossing points respectively. We can adjust the yew angle of the foot and pitch angle of the sensor to
simulate walking behavior as shown in Figure 7. The data is collected using Labview program at a sampling rate of 1
kHz. The raw data is then going through a preprocessing procedure as shown in Figure 8. First of all, we use a low pass
filter with a cutoff frequency at 100Hz to remove high frequency noise. The filtered data is then normalized with respect
to its initial DC value to obtain its transmissibility. Ideally, the normalized signal will start from 1 which represents no
light loss without any loading on the surface of sensor. When there is a loading acting on the sensing surface, the
corresponding channels will have less light due to bend loss which means the transmissibility is less than 1. We can
obtain the signal changes in each sensing point by adding signals from the crossing fibers and representing them into a
2D color image.

Proc. of SPIE Vol. 8348 83481U-5

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 10/15/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

 Figure 6. Foot orientation and fiber layout of top and bottom layer.

(a) (b)
Figure 7. Rotating prosthetic foot to simulate walking direction (a).Yew and (b).Pitch

Figure 8 Flow chart of signal processing.

5. RESULTS
A series of cyclic load test had been conducted by using the foot pressing at different sensor location. We are
performing 10 cycles of each foot shape with the heel center pressing at various positions along the BY4 (Figure 6) at

Proc. of SPIE Vol. 8348 83481U-6

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 10/15/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

 different yew angle (from left 15o to right 15o) and pitch angle (0, 2.5o and 5o). The loading rate is 6mm/sec and the hold
time at the maximum loading is around 5 seconds.
Figure 9 shows the filtering result of a raw data at the bottom layer as described in previous section. Filtering cleans up
most of the environmental and experimental noise. We can also observe some fibers have higher response (BX4 and
BX5) then the others (BX7). The change of transmissibility will be more obvious when we normalized the signal as
shown in Figure 10.

(b) Raw data (a) Filtered data

0.8 0.8
BX0
BX1
0.7 0.7 BX2
BX3
Amplitude of Intensity (V)

BX4
0.6 0.6 BX5
BX6
BX7
0.5 0.5

0.4 0.4

0.3 0.3

0.2 0.2

0.1 0.1
0 5 10 15 20 25 30 0 5 10 15 20 25 30

Time (sec) Time (sec)

Figure 9 Filtering of BX optical data: (a) Raw data, (b) Filtered data.

1.05

BX0
BX1
Transmissibility (%)

1
BX2
BX3
BX4
BX5
0.95
BX6
BX7

0.9

0.85

0.8
0 5 10 15 20 25 30

Time (sec)
Figure 10 Normalized BX optical data.

Proc. of SPIE Vol. 8348 83481U-7

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 10/15/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

 Figure 11 and 12 shows the responses of a hammer toe foot with heel center pressing at the crossing point of BX4 and
BY4 at 2 different walking direction. Figure 11 walk at 15o to the left and Figure 12 walk at 15o to the right. A special
designed foot distribute the weight only at the heel and ball area as shown in Figure 5(c). The blue color represent less
light transmit through fiber due to bend loss which corresponding to higher stress area. We can observe TX4 and TY3 at
the top layer have less transmissibility which is the corresponding position of the heel center. The bottom layer give us
information of the walking direction with the Figure 11 gives less transmissibility at BY2 and Figure 12 gives less
transmissibility at BY5. This kind of pattern change at different layer gives an opportunity for neural network to classify
their walking direction.

Top Layer Botom Layer

1 1
0 0

0.95 1 0.95
1

2
2
0.9 0.9
3
TX

BX
3
4
0.85 0.85
4
5

5 0.8 0.8
6

6 7
0.75 0.75
0 1 2 3 4 5 6 0 1 2 3 4 5 6 0
TY BY

(a) (b)
Figure 11 The (a) top and (b) bottom layer responses of a special prosthetic foot with 15o walking direction to left.
Top Layer Botom Layer
1 1
0 0

0.95 1 0.95
1

2
2
0.9 0.9
3
TX

BX

3
4
0.85 0.85
4
5

5 0.8 0.8
6

6 7
0.75 0.75
0 1 2 3 4 5 6 0 1 2 3 4 5 6 0
TY BY

(a) (b)
Figure 12 The (a) top and (b) bottom layer responses of a special prosthetic foot with 15o walking direction to right.

6. CONCLUSIONS
The ultimate goal is to build a planar sensor which is able to recognize the foot shape, its walking direction and
pressure/shear stresses acting on the sensor. The uniqueness of this design is that we are able to change the yew and
pitch angle to simulate the walking behavior. The grid design of our optical sensor can catch large area of stress
distribution with good resolution. In this study, we use TSL260RD photo detector to improve the coupling of our fiber.
We also build 3 different prosthetic feet for simulating normal, flat foot and unhealthy foot. We also coded an algorithm
to preprocessing raw data into image pattern information which can be used as the inputs for neural network. Our next

Proc. of SPIE Vol. 8348 83481U-8

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 10/15/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx

 stage is to further characterize the sensor against shear forces and to train the sensor to recognize both normal and shear
force, shape of the three different full scale prosthetic feet, and also the location where the load is applied using neural
network algorithm.

ACKNOWLEDGEMENT

This work was supported in part by the VA Rehabilitation Research and Development for Limb Loss Prosthetic
Engineering. Service, grant A0806C.

REFERENCE
[1] K. Sundara-Rajan, G.I. Rowe, A.J. Simon, G.K. Klute, W.R. Ledoux, A.V. Mamishev, “Shear sensor for lower
limb prosthetic application,” Proc. BSEC, 1-4 (2009)
[2] S. Karki, J. Lekkala, H. Kuokkanen, J. Halttunen, Development of a piezoelectric polymer film sensor for plantar
normal and shear stress measurements, Sensors and Actuators A: Physical, Volume 154, Issue 1, 31 August 2009,
Pages 57-64, ISSN 0924-4247, DOI: 10.1016/j.sna.2009.07.010.
[3] C.Y. Huang, W. Ledoux, W.J. Wu, W.C. Wang, “Composite fiberoptic distributive shear/pressure sensor,” IEEE
Sensors Journal , Vol. 7, No. 11, p.1554-1565, 2007.
[4] W.C. Wang, W.Y. Hunag, W. Ledoux, P. Reinhall, “Developments of a Microfabricated optical bend loss sensor for
distributive pressure measurement,” IEEE Trans. Biomedical Eng., Vol. 55 No.2, p.614-625, 2008.
[5] W.C. Wang, R. Panergo, C.M. Galvanin, W. Ledoux, B Sangeorzan and Per G. Reinhall, "A flexible micromachined optical
sensor for simultaneous measurement of pressure and shear force distribution on foot", Proc. SPIE 5047, 275, 2003.
[6] W.C. Wang, Ledoux W.R., Huang C.Y., Huang C.S., Klute G.K., Reinhall P.G., “Development of a microfabricated
optical bend loss sensor for distributive pressure measurement,” IEEE Trans Biomed Eng., 55(2), 614-25, 2008.
[7] W. Soetanto, N. T. Nguyen and W.C. Wang, "Fiber optic plantar pressure/shear sensor", Proc. SPIE 7984, 79840Z,
2011.
[8] J.H. Ho, W.C. Wang, “Development of a polymeric capacitive 3-D tactile sensor,” 17th SPIE NDE health
monitoring and Diagnostics, San Diego, CA., 7 - 11 March 2010, SSN10-86
[9] Y.-L. Chou, S.-S. Shi, G.-F. Huang, and T.-S. Lin, "Interface pressure and gait analysis in different walking speeds
and on the below-knee amputees with multiple axis foot prosthesis", Biomed. Eng.: Appl. Basis Commun. 15 (5),
pp. 38–42, 2003.
[10] J. Missinne , "Flexible miniature shear sensors for prosthetics", SPIE Newsroom. DOI: 10.1117/2.1201104.003664,
27 April 2011,
[11] C.Y. Huang, P. Reinhall, W.J. Wu, W.C. Wang, “A wireless composite optical bend loss sensor for pressure and
shear measurement,” SPIE NDE health monitoring and Diagnostics, San Diego, CA., SPIE6177, 2006,
[12] C.S. Liu, W.C. Wang, “Composite fiberoptic sensor for shear and plantar pressure measurement,” SPIE NDE health
monitoring and Diagnostics, San Diego, CA., SPIE 6532, page 65431k-1 to 10, 2007.

Proc. of SPIE Vol. 8348 83481U-9

Downloaded From: http://proceedings.spiedigitallibrary.org/ on 10/15/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx