Good afternoon everybody.

Again my name is root and the subject of
my study was called the calibration of

an insecure pressure measurement system
using different interface materials and

was conducted by myself along with some
assistance from Robert kissed the bird and

musical glare.

So in the fitting of a prosthesis
to an amputee the socket is

usually considered one of the most
important components in that process.

He says because it's the one component
that's actually interacting directly with

the residual limb and so there are certain
parts of the anatomy where the pressure is

more sensitive than other areas and
there are other parts there and

I mean that is more
tolerable to pressure and

knowing these areas is very important to
how we fit our stock into the patients.

If you look here in this diagram
we can see the trans to the N.P.T.

Here are the places that are more
tolerable to say to pressure things such

as the medial to Bill flares and
along with the teller tendon with you look

at the other pressure sensitive areas or
more the bony prominences such

as the distal end of the tibia and
the feet below the head.

So if you don't get these things right.

You going to have poor socket fit and

they are going to cause pain discomfort
to the patient as well as possible skin

condition problems such as
I'll Servatius that could occur and

overall you might reduce some
of the stability in your socket.

So currently in clinical practice we
evaluate on prosthetic fit based on a lot

of qualitative measures such as the
practitioners personal past experience or

perhaps the feedback from
the patient as well as using

visual cues such as
the coloration of the skin.

In a study in two thousand and
he mentioned that it's important for

practitioners to move away from this
kind of qualitative national and

system something that's more quantitative
so that we can obtain better understanding

of interface pressures have an objective
evaluation of prosthetic design

in function and hopefully in the future
we can advance our socket designs and

have a better understanding
of prosthetic biomechanics.

So one such system that allows for
the quantity of measurement of pressure

systems inside the socket is the tech
scans of socket system it's in

socket pressure mapping that is based off
of four sensing resistive technology or

at us are and basically you have.

As you have two very thin milers sheets
that sandwiched in between pressure

sensitive resist of AIM and the sink is
arranged in sort of a matrix form and

wherever the rows and
columns intersect and

creates a sensitive area they can detect
pressure and they call these sand cells.

So when you apply pressure
to the sense cell a century.

You're creating a change in resistance

that results in a change
pressure that we see.

For the ninety eight eleven A.M.
model for Texas and

they have ninety six individual sense
cells for detecting pressure and

the next rating is about twenty five
P.S.I.I. or hundred seventy point far for

killer Pascals this system only
measures normal forces not sure.

So according to tech scans manual they
say the calibration should be done

best to mimic whatever environment
you're going to be testing in and

some considerations you need to look at
are the temperature that you need testing

in the surface curvature how long you'd
be collecting your data on the pressure

that you're applying and as well.

The material interface and
I want to focus mostly on the material and

from here is a diagram of a cross-section
of a socket sensor and if you look here.

The areas that are more rounded that's
actually where the intersection between

the rows and columns occurs and that's
where the four sensing resistance and

gives the depression areas here
that's then an active area so

it doesn't actually sense
pressure it's like a valley.

So if you had an infinitely
complex material that was

applied to the surface of the sensor and

you apply the load to it essentially
that material can follow the pattern of

your sensor sheet and fill in the gaps
in the hills and valleys in your sensor.

And so with this you can see that some
of the load is experience in the active

sensing areas and some of the load is
also experienced in that in active areas.

Not be taken to religious smartly
compliant and you apply a load to

it you can see that the material also
conforms slightly to the hills and

valleys of the sensor.

But it doesn't fill in quite as much as
an active area as the infinitely compliant

material does so with this condition.

You have more force of acting on the
active areas as opposed to the infinitely

compliant stage.

Now you take this to the next level.

You look at a completely non
compliant material basically it's

bridging across the entire
sensor in the even.

Actually an active areas.

So all the load is essentially
on to the active sensing areas.

So there directly from this
diagram you can see that for

this condition you would have higher load
output or measurements from the sensor

using a non-compliant material
versus a blind material.

In the lot of the studies that
I read up on using the sensors

a lot of the calibration that
were performed performed without

using the material interface
to be tested in later on.

So essentially they calibrated
the sensor without a material and

then they stuck inside some one socket
that may have used the liner or

prosthetic sock or some other material
interface in one thousand nine hundred

eight why did a study on using ensure
pressure measurement system and

he found that the sensor
output highly depended on.

The contacts surface hardness and
so I wanted to evaluate this

using it in shoot in socket pressure
measurement system using sockets Cers.

So the first step to
valuating the differences in

calibration with materials was to first
fabricate and validate a calibrated that

could accommodate not only of
the dimensions of the F. socket sensors.

But could also accommodate to
the different material thicknesses that

I would be testing the pressure
system also needs to be able to apply

a uniform pressure across the sensor and
had said within at least one hundred

seventy two point four killer Pascals
which is the max and rating of the sensor.

So for the calibration I want to be able
to determine what the fact of the pressure

output would have been if I calibrated
with the material without the material.

And so I want to quantify the effect for
the following prosthetic materials

that are used very commonly in such
as prosthetic socks liners silicone

liners and polyethylene form
probably no purpose is purely from.

So here's a mock up of my test equipment
that I used as Actually I have a top board

in the bottom for
the air comes in to here and

inflates the pressure bladder
Superman of your thing and

the pressure is measured using the gauge
when your thing bladder inflates it

compresses on the material where
the sensor replace either on top or below.

And over here we have various height
plastic spacers that are used an integer.

Depending on how thick my
material is going to be.

To validate the calibrator I also wanted
to make sure that across the board where

the sensor was going to be placed that
the pressure was going to be even so

I stuck four different guys
a resistance transducers on the bottom

of the board one of
the Center at the corner

on different edges to see if the pressure
distribution was even across the board.

So I bought it calibrated using
the four resistors are transducers at

four different levels of pressure
at zero twenty point seven

thirty four point five forty eight point
three kill of ask Els and I compare that

to the input pressure that was supplied
by the air compressor using a sample T.

test with P.
less than zero point zero five.

In terms of checking the differences
in calibration with the material

without a material I did everything at the
same pressure level of one hundred three

point four killer Pascals which
translates to fifteen P.S. I.

And I have three separate conditions the
first condition is my control condition

and here I am calibrating
the sensor with no material and

then test them in also with no material
and so this is sort of the baseline for

comparison across the other conditions.

The second thing I did was I compared
calibration of the sensor with no material

and then following that I'm testing
the same sensor using the material and

I broke that down to measuring
two different things.

I put the sensor both on
top of the material and

I also checked the sensor when I was
below the material and this is sort of

simulating a situation if you're
putting the sensor between the liner in

the skin versus if you're putting the
sensor between the socket and the liner.

The final condition that I tested was
calibrating the sensor with the material

interface of interest and
testing with the same material and so

I'm expecting here to have more accurate
results are similar to the control because

you're actually calibrating to the same
environment as your testing and again as

with the sub with the sensor being above
the material and also below the material.

And collecting the data for
three trials of fifty herds and then I

compare the pressure values to the control
using the sample to test again and.

So here are the results of a team from the
validation of my high calibrator on the Y.

axis you see the pressure and kill of
Pascals on the X. axis you can see.

With the color bars the four different
transducers there were on the board board

and you can see at the zero Pascal level.

When I put it.

Zero kilo Pascals as in such a weight and
put any pressure in my pressure

sensors were giving me outputs
there are lower than zero and

they're all consistently lower than zero
Same thing with all the other pressure is

what I put in twenty point seven thirty
four point five to forty eight point

three as an input pressure all four of
my sensors were still reading lower

than the pressure that was imported and it
was consistent in the same four sensors.

So looking at the values of the
differences between the input pressure and

the pressure detected by
the friends do series.

You can see on average
the pressure measurement for

the four sensors was roughly
one point one kill of ask for

the input pressure with a standard
deviation of point four eight.

Here are the results from the calibration
experiments that I did again on the Y.

axis you have your pressure value
in Cope ask Els and on the X.

axis you have the control vs the different
types of materials that I tested so

breaking up the issue here.

The four bars here the first two bars
are the material commission a where I

calibrated without material interface and
then I tested it with material and

the last two bars the green in the purple
are the conditions where I calibrated and

tested with the same material interface
the first bar being the sensor above in

the second bar being the sensor
below the material is interested in.

So I am putting Again as I mention
one hundred in three point four color

Pascals and that's the line that was
the kind of the target to go for

but the control I ended up getting only
one hundred one point five killer Pascals

which is under reading about to kill a
Pascal's from the can from the input which

is kind of consistent to what I saw when I
validated my calibrated it in the Reading

a little bit lower than expected for the
other materials you can see that if you

calibrate without the material interface
you're looking at the tend to have lower

pressure values and if you calibrate it
with material that you investigating and

for their own for the most part the.

One that had the same statistical
differences were the ones that had

calibration with no material with
the sensor below and you can see that with

the asterix for the gel three millimeter
the gel six millimeter the P.

light and the silicone the P.I. was
the one thing the kind of stuck out for

me didn't really matter
if you calibrated with or

without material it seemed like it was
long as the sensor was below the material.

It had very high output for sense for

the sensors which were about one
hundred points above till past bells.

So it seemed like my control was lower
than my input pressure which was again as

I mentioned consistent to what I saw
with the validation of my calibrator

this may have been as a result of
the way the the pressure transfer

of the pressure gauge was read.

There is an inherent poster
minus point five P.S.

I error within the pressure gauges and
their read and

a lot of digital and so there might
have been some reading errors for that.

And possibly from the inputs of the output
there is some air loss as well.

We do find the US moving in difference
was showing the pressure output for

calibration when there was no material and
the sensor was tested below the material

and this made them because of the
difference in the material stiffness you

have in this clip calibration environment
vs what you have from your test garment.

In terms of comparison between
the materials themselves I think the drama

between the materials are very similar
to each other so we didn't really see

too many significances between that except
for the for the pure light liner the show

kind of inconsistent results studied by
June in one thousand nine hundred eight.

Also found that when they tested
a transducer using pure light they had

overshot some of the pressure
values that they had and

so the also had very one over estimation
of how much pressure was going on.

And so this might have been because
of the income press ability and

the stiffness of the material as I
mentioned earlier when you have a hard

surface or hard incompressible
material that's noncompliant you

tend to have higher pressure values
because it's not filling in the in.

The nonsense it is in
Africa is near sensor.

So some of the limitations in
the study the pressure gauges and

mention was analog and
had an inherent error to it.

The materials that I use may not have
had a perfectly consistent uniform

distribution because of the material
they get is not going to be perfect and

the equipment that I had I put the
pressure transducers on the top surface of

the bottom plate and
they may not have been exactly flush

trying to get the most part as I could but
there's always going to be armed factions

with that I also examined one
pressure level at the fifteen P.S.

I with might be interesting to see what
differences you get if you have higher or

lower primaries and the drain in
the loading I noticed that when you

lower the pressure sensors for a certain
amount of time if you kept lowering it.

The pressure would rise until as
they did offer a certain point.

So there is some drift going on and
this was also noticed by Poliakoff and

Texaco manuals and so
they recommend that you hold it for

at least thirty to sixty seconds.

But when you were to material in there.

Sometimes you get differences in how long
at the old before you get a consistent

result.

And there's no real data out
there in terms of how long and

how much you can use the sensors for
so for

to avoid what we can't really tell because
we don't know how long they're used.

So in the future be kind of interesting
the salmon again as I mentioned different

pressure levels to see if there's
a difference if you have higher or

lower pressures be interesting to see
also in the more clinical setting like

in the socket to look at curve surfaces
and a dynamic and condition when your

load is sickly so you can look at
history says and professor that.

So in conclusion when you're calibrating
the sensor without a material and

then you test it with the material and

you have the sensor below you going to get
a test sophistic Lee different results.

So it's recommended in any case
to calibrate it with material as.

The same if you're that going to be
using a test environment for accuracy.

I like to think the following people will
talk with Williams who works at tech scan

for providing me the sensors and forgive
me of the advice on how to use them.

Mark pickin for his advice and
how he's using clinic to get a call blared

Robert his number for
there is a provision during this project.

ROBERT down for not providing me
with some of the materials and

that information on the materials.

Dr Stevens beagle Jonathan jars and
the rest of the kitty a staff for

letting me use their facilities
to build my calibrator.

And to give me advice on pressure sensors
and finally to duck if he says no for

system that is tickle advice.

Is my reference or any questions.

Yes.

Part of this research
was possibly he's right.

So you know you want when you
get your money he speculated

boy boy it's whether the sensors
would be placed correctly.

Yes You know they start

right now as I mentioned with this
there are many factors that affect

the calibration accuracy of this things
like temperature on the surface curvature.

And as well.

The material interface Ideally I'd like
to see this being applied where they

can calibrate directly within the socket
and that way you take care of

all the variables all at once for right
now we don't have the information for

all these little pieces in terms of
how they affect the calibration.

So that's kind of why I'm doing this
research is that I have a component of

that information so ideally I kind of like
the way they do in shoe measurements where

they put sensors directly underneath
the installs of their shoes and they

can put their body weight on it you can
load that and using the information from

the person's body weight and knowing all
the commissions that are going to be used

in testing you can get much more accurate
results and calibrate and so in the end.

Hopefully you can do the same thing
with this only instead of doing

it with a calibrating you can do it
within the socket using their weight.

As a role in the.

Work.

You're right.

Right.

And.

I do believe the both components
are relevant in terms of

affecting the residual limb and
terms of pain and discomfort so

share is definitely a component
own terms of skin breakdown but

I believe that when you have
something digging into you.

That's going to be irritating as well.

And right now we don't have actual values
to know how much we can tolerate in terms

of that pressure for your skin in
forevers soft tissue muscle fat

all those because there's confined for
a patient to know how much we can endure.

We don't have data on that.

So with we can gather data using
pressure systems like this to know.

Well the Hertz of this P.S.I.I. for this
type of body composition we can use that

to help guide for our pressure off
our stock of designs in the future.

And.

Right.

Yeah.

But.

Yes.