Force/time curves comparison and interpretation

Eric

Thanks for your thoughts

I agree that you have clarified the problem nicely.

Just wanted to make sure we are talking about the same thing when I talk about edge effect.

I seem to remember (but cannot quote details) that matrix force sensors will give reading to cells peripheral to those actually in contact with surface of the load. Depending on sensor size this (because the periphery will significantly surface area of contact) has potential to influence of FTCs (larger cells will give greater error?).

I think I understand your idea with the bladder but if I am understanding you correctly would need to create some load bearing templates with different radii to evaluate this, simply placing entire sensor under bladder will not give the answer to this question.

I think this is a neat idea and when I get my bladder set up right (not an incontinence issue) I’d like to try this.

On your second point let me try and break down my thoughts and see if you can find problem with my reasoning.

If you place the sensor between foot orthoses and shoe the interface will be flat, there should be no crinkling if care is taken with fit into shoe.

We can capture data for several steps and look at Total foot FTCs.

My initial idea was to define the shape using “shape factor” and look at variance etc to evaluate our concerns regarding interstep variability but using Simons mantra of “KISS” let’s just make some qualitative judgments from shape, min and max force values and midstance timing.

Also ignore or compensate for “edge effect” perhaps in the way you mentioned.

Now using same sensor, calibration data, shoe, foot orthoses, foot, measure the FTC at foot orthoses / foot interface.

The resulting FTC should be same except for influence of;

1 Crinkling

2 Timing lag period caused by Visco-elastic properties of foot orthoses material

3 Temperature gradient

Therefore we can study effect, at least on FTC, of the variable of “crinkling” +- effects of 2 and 3 which I would estimate would likely be constants for foot, orthoses and sensor.

If you cannot find flaws in my reasoning I’d like to put this to the test when I get a bit of spare time.

Look forward to comments on this


Cheers

Martin

The St. James Foot Clinic
1749 Portage Ave.
Winnipeg
Manitoba
R3J 0E6
Phone [204] 837 FOOT (3668)
Fax [204] 774 9918
www.winnipegfootclinic.com

 

Simon

thanks for your suggestions. I am thinking about this and doing some reading to understand visco-elasticity right now, once I've got a more in depth handle on this I'll pick up on this.

thanks for pointer to V elasticity it more interesting and important than I had realised


cheers

Martin

sThe St. James Foot Clinic
1749 Portage Ave.
Winnipeg
Manitoba
R3J 0E6
phone [204] 837 FOOT (3668)
fax [204] 774 9918
www.winnipegfootclinic.com

 

I'm not sure what you mean by edge effect.

The load bearing templates could be different sizes of orthotics. For example take an othottic for a size 15 foot, size 12, size 10, etc.

I'm not quite sure what your shape factor is. I'm also not sure what you want to compare it to. Before you start collecting data you should have a research question like pointy force time curves are associated with hemoroids. You can look at your data and either keep or discard your hypothesis.

Regards,

Eric

 

" edge effect" is when the force applied to a cell causes a force to be measured on an adjacent cell which is either not loaded or partially loaded and not truly representing the vertical vector of the GRF.


Shape factor is a mathmatical representation of a FTC which can be defined from parameters defined in the paper cited at start of thread.



They are applicable to all geometric shapes.
They are independent of scale and orientation.
They are dimensionless numbers.
They are not overly dependent on one or two extreme points in the shape.



the hypothesis is that the shape factor of the FTC measured between the foot orthoses and the shoe is not significantly different from the shape factor of the FTC measured between the foot orthoses and foot.


This may be useful because I think it might help identify the degree of artifact which may occur because of distortion of the matrix sensor when placed on surface of foot orthoses particularly at plantar calcaneal area and also cutway window to first metatarso-phalangeal joint either because of "edge effects" or "crinkle" effects.

With the fairly rigid material FScan sensors are made from, in my experience, although they are very thin, they do not conform well to steep contours. Because the foot orthoses / shoe interface is flat (or at least can be selected to be flat) these artifacts from distortion of the sensor are unlikely.


Currently to my knowledge this have never been investigated.


I seem to be out on a limb here, perhaps I should take a few measurements and if they look interesting enough revisit this on the forum



cheers



Martin


The St. James Foot Clinic
1749 Portage Ave.
Winnipeg
Manitoba
R3J 0E6
phone [204] 837 FOOT (3668)
fax [204] 774 9918
www.winnipegfootclinic.com

 

Hi all,

I have been looking into Shape Factor and I think my take on it might be a bit different to Martin's (then again, its hard to understand and we may be talking about the same thing).

In the 1997 paper 'A theory of metabolic costs for bipedal gaits' by Minetti and Alexander, they state that as we increase speed in gait, three predictable things happen.
1.We increase our stride length
2.Duty factor reduces. That is, the fraction of time that each foot is on the ground per stride. Essentially, walking duty factor will be more than 0.5 (usually around 0.6) and running duty factor will be less than 0.5 (usually around 0.35).
3.Shape factor increases in walking from a slow walk (0.2) to a fast walk (0.7). Shape factor falls abruptly in running to around -0.1. The original paper to describe this is Alexander, RM and Jayes, AS. 1980. Fourier analysis of forces exerted in walking and running. J Biomechan. 12:383-390.
Shape factor is defined as a ratio of Fourier coefficients that describes the time course of the forces exerted on the ground. I emailed prof Alexander and he further explained that "shape factor describes the shape of a force plate record of the vertical component of the force on a foot, during a single step. It is zero if the record is like a single peak of a sine curve, positive if the record is flatter topper or two peaked (as in walking) and negative if it is bell-shaped (as in running)".

In spite of that I admit I still don't really understand it.

In the 1980 paper, Alexander states in the abstract that "The forces exerted on the ground by men walking and running have been recorded by means of a force platform (it was a Kistler plate). A simple method of Fourier analysis has been used to analyse the records. It shows differences between walking and running, between slow and fast walking, between accelerated and decelerated walking and between different individuals walking at the same speed".
Just in case anyone is wondering, Fourier analysis is the technique of transforming a complex waveform into its sinusoidal components (I guess this means regular waveforms). Fourier was the guy who came up with the idea in trying to understand the heat equation. So in regard to the forces in gait, like the example of Fouriers first examination of heat, the forces produced could be described as complex waveforms that can be broken down into separate regular (sinusoidal) waves of different frequencies.

I really don't know much about measuring forces in gait as I have never used a forceplate or in-shoe setup, but I wondered if this Fourier analysis and Shape factor is being used today.

I looked on the Tekscan website and in explaining the add-on feature of CoM'nalysis, the introduction states that sophisticated Fourier and Harmonic analysis (are they the same thing?) is used to come up with four Indices:
Purity Index - represents the efficiency of potential energy to kinetic energy transfer.
Symmetry Index - represents the degree of symmetry between the left and right foot
Energy Efficiency Index - represents the energy conserved in moving the CoM against gravity
Gait Perfection Index - combines the Purity, Symmetry and Energy Efficiency Indices.

Is one of these indices the same as shape factor?

Rebecca

PS: I have PDFs for some papers of Alexanders which discuss shape factor. I don't know how to put the PDF in my post but can email.

Alexander RM. 1989. Optimisation and gaits in the locomation of vertebrates
Alexander, RM. and Jayes, AS. 1980. Fourier analysis of forces exerted in walking and running.
Alexander RM. 1991. Energy-saving mechanisms in walking and running.
Alexander, RM. 1984. The gaits of bipedal and quadrupedal animals.

 

Rebecca if you could please email me a copy of the Fourier analysis I'd be grateful, I have been unable to otherwise get a copy. Good for you contacting the root of the papers.


the COM and purity index we already discussed a bit in Simon's original post on measuring PCIs and kind of dismissed.


since this thread is getting a bit messy I just want to recap a bit.



Use of shape factor was included in creating Alexanders model which was cited in Kevins thread on Gait efficiency (he posted the paper) which is where I picked up this thread in an attempt to understand its meaning and possibly its use for other purposes such as looking at sensor artifact.

Simon suggested simplifying the analysis of the curves by breaking the curves down into the linearly related parts of curves, concentrating on the slopes.

I thought this a good idea, however I think that the midstance part of the curve is often the most disturbed part when function is being explored and not that linear. Which is why it seems likely to me dividing the total FTC into forefoot and rearfoot FTCs may be more revealing but we have not discussed this.

Simon also suggested understanding the viscoelastic qualities exibited during stance and how this might effect the shape of the curve, I am still reading around this, can understand why he suggested it but need more time delving before commenting any further.

Craig P posted some before and after FO FTCs measured with inshoe sensors.

I found ambiguity in trying to decide if the curves looked significantly different.

I wanted to look at what we could usefully gleen from a total foot contact FTC, how we could make meaningful comparisons between them, also importantly if it would be expected to appear different according to sensor postion in the ground/shoe/ foot orthoses / foot interface for the same subject/shoe/ foot orthoses /walking speed.

No one seems to have contradicted my suggestion the the FTC shape should look pretty much the same now matter which layer the sensor is sandwiched between (even though the pressure distribution will be profoundly different for each of the interfaces)

The thread then kind of morphed into the possibility of understanding the behaviour of in shoe matrix sensors. Eric has suggested alternative approach, defined the problem clearly but I dont think I have explained my idea to him clearly so far.

As usual the focus has shifted around a bit.

In a couple of weeks I wil have some time to try out my idea if it still seems reasonable.



Perhaps there is still a bit of mileage on this yet.

did your math pal make any sense of this?


cheers

Martin


The St. James Foot Clinic
1749 Portage Ave.
Winnipeg
Manitoba
R3J 0E6
phone [204] 837 FOOT (3668)
fax [204] 774 9918
www.winnipegfootclinic.com

 

Sorry Martin, she has just moved house, now lives 400km away and there's a delay with Telstra hooking the phone line up, plus the start of the new school year - I'll let you know as soon as.

Rebecca

 

Martin:

Just an observation from a grown son of a former English teacher.....have you ever tried stringing three to four sentences in a row into the same paragraph?? This standard writing technique would make your postings not seem so long to those of us trying to follow along.:drinks

 

Thanks for tip Kevin. the irony is that I thought by splitting ideas into discreate chunks it would make it more understandable. I'll pass your comments on to my wife who claims that I am illiterate (she is a writer) and gets particularly distressed when I beat her at scrabble which is not that often and I recon less to do with literacy and more to do with braille (also called fingertip in the bag cheating)

cheers

Martin


Martin


The St. James Foot Clinic
1749 Portage Ave.
Winnipeg
Manitoba
R3J 0E6
phone [204] 837 FOOT (3668)
fax [204] 774 9918
www.winnipegfootclinic.com

 

Bruce

I had a fairly quick scan over this thread so I hope someone has not already covered this.

You wrote

We would get perfectly useful data if the data one required was about total force applied to the foot and not discreet force/area data. If one was interested in how an orthosis changed the gait interms of forces and moments and powers then an inshoe, foot to FFO interface, pressure system would be inappropriate. Horses for courses one might say.

My biggest bug bare about inshoe systems ie insoles is that there is no reliable reference frame. IE when you use a pressure mat or force plate the vertical y axis is in the global reference frame of the lab and the orentation is the floor. The viewing reference for a pressure mat system is in the plane of the computer screen, which we can easily visualise in terms of the global reference. However when using insoles the orientation of vertical applied discreet force is unknown and yet it is displayed on the same flat computer screen, which we then visualise intuitively in the global reference frame.
This may be ok as a comparative reference for discreet force area applied to the foot but even then is a confounding element.
If one wanted to know something about the moments applied to a certain joint within the foot eg the 1st met cuneiform, how could you calculate this from the force /area data from an insole system. Moment = (force * Length) * cos or sin theta. However there is no way to know direction of force data from the insole.

Do you think this is a valid view?

Cheers Dave

 

I see your point Dave. The sensor measures force normal to the sensel, yet the angle of the sensel to the floor may be changing. With regard to calculating moments, I really don't think a pressure system such as F-scan is the way to go since you really need the 3D vector.

 

Martin:

Don't worry. You aren't the only one that I have "spoken to" about their writing here on Podiatry Arena. Since many of us only know each other here on Podiatry Arena by our postings, we can often only judge the intelligence of others on this forum by the quality of postings, which include content, intelligibility, grammar, punctuation and writing style. Therefore, those individuals who take the time to write with a logical style, good grammar, correct punctuation and using good writing methods, will always be perceived by others to be more intelligent on a forum such as this since their writing is all we have to go by in assessing their knowledge. Those that write carelessly, don't proofread their postings and/or use cryptic or confusing language tend to be viewed as either not being as intelligent or not caring about how they "appear" to others. This is just the nature of how we humans judge each other when we have nothing to go on but the written word.

I'm not picking only on Martin since his postings have excellent content. I hope this message of mine serves its purpose of making those of you who choose to contribute to this international podiatric educational forum make certain that your postings, that are being read by hundreds or thousands of podiatrists and other foot-health professionals around the world, reflect the type of image you want others to have of you. Of course, if you don't give a hoot about what your colleagues think about you, then just continue to write to this forum with no regard to the quality of your postings.

 

Fourier analysis paper posted on behalf of Rebecca.

The way to think about this is in terms of a musical chord. A chord played on a guitar is made up of a number of individual notes being played together. When we add the waveforms of each of the individual notes together we get the waveform for the chord- this process is called Fourier synthesis. Similarly if we start with the waveform for the chord we can extract the waveforms for each of the notes that make up the chord- this is Fourier analysis.

 

Simon

You wrote

Yes quite so, Horses for courses as I said. Trouble is, I think people use and view pressure mats and in shoe data interchangeably when, in my view, they can have very different raison d’etre.

For instance, since the foot / orthosis interface is not flat like the floor foot interface the insole system will give spurious and incomplete 3D Force/area data presented as vertical Y axis 2D force/area data. IE it will collect data about horizontal forces and present it as vertical force data, which it is relative to the interface but not to any known reference frame. How then can it produce a reliable CoP progression.
Thought experiment: Draw a "straight" line around a sphere. Flatten the sphere into a 2D plane and is the line still "straight"? Can one draw a straight line on a curved surface?
We know the geometry of a sphere, eg like the earth, and so we can make maps from which we can navigate with precision. However we do not know the geometry of a certain orthosis and how it changes its geometry in shoe during gait.
Now draw a line representing the ideal CoP progression on the surface of your orthosis and then unfold the orthosis onto a flat surface. Similar to what the software is doing when it presents the data for viewing. How will the CoP progression look now? The horizontal distance between two fixed points would appear much longer on a flat diagram than on the actual orthosis.

If we take four equidistant sensor cells, put them on a flat surface and apply a vertical force to each, it would be quite simple to work out the resultant force/unit area and the CoP. If we then applied a force at an angle to each sensor we could still work out the vertical force/unit area because the sensors only measure perpendicular force and not shear. We could work out the CoP in terms of vertical force only. Imagine now the sensors are on each vertical face of a cube and a perpendicular force (from the inside out) is applied to each, where is the centre of pressure? There isn’t one, however now fold the cube flat so that each of the vertical faces are now horizontal and equidistant from each other. Now, using the same data, there is a CoP, which is somewhere in the centre face of the now flat cube. This appears to be what an insole system attempts to do and presents to the viewer? Doesn’t seem right to me.

I don’t know much about the algorithms employed within a proprietary in sole pressure system but I would be interested to know how it is possible to extrapolate CoP, in terms of a flat surface, from an unknown flexible and flexing, multi curved surface.

Cheers Dave

 

:good:
I think this is leading toward non-Euclidean geometry amd Riemann geometry which is where my brain starts to melt- http://en.wikipedia.org/wiki/Riemann_surface

 

Thanks, that's very helpful!

 

Creating an x/y coordinate for an in shoe sensor is quite easy . All you have to do is assume the sensor is flat and plug the numbers into the formula to get a value for center of pressure.

This points out the problem I've been having with this thread. We are taking data and plugging it into formulas without really examining what the results should mean. Why should the shape of the force time curve be relevant to clinical outcomes? Simon mentioned about how his Groucho walk showed the "best" force time curve of several different gaits. Should we all walk around like Groucho?

I believe that in shoe pressure analysis can answer important questions, however we should apply the right kind of data to the right kind of question. For example, you have just modified a neuropahtic patient's insole and you want to see if you got pressure reduction. Center of pressure is not relevant to that question. The value of pressure around the point of interest is important. There are other questions and other variables that would apply ot those questions and we should use be aware of why we think a variable is important.

Cheers,

Eric

 

:good:

:good:

A force plate, pressure plate and inshoe system all measure different things... but some use the data interchangeably as if they are one and the same. This includes some companies that are marketing these systems...
On top of that, it is easy to get caught up in the 'sexiness' of technology, and there are a lot of people who read too much into these measurements. Not so much of a problem with the contributors to a forum like this, but I am sure we have all rolled our eyes when a patient presents a computer read out from a clinic or a sports store which explains to them what their problem is... and they have not even had their feet touched :bang:

 

As I said earlier:

Here's your starter for ten: The gradiants of the force time curve will provide information regarding rate of loading. If the rate of loading is important in clinical outcomes as intimated here: http://www.podiatry-arena.com/podiatry-forum/showpost.php?p=31436&postcount=101, then this may be important in terms of clinical outcomes.

What else might we glean from the force/ time curve?

 

David;
I look at this as a research vs clinical debate.

Kistler plates are invaluable in calculating joint moments when paired with a 3D kinematic capturing system. I can somewhat appreciate that the math would be easier to calculate if the data were generated from a mat or plate on the floor. They would work great for calculating forces, moments and power at specific joints when patients walk barefoot. Once you add a shoe you blunt the plates data and often the 3D capture data as well.

From a clinical standpoint, such as mine, in-shoe pressure systems are much more affordable and practical on a day-to-day basis. I have not had the luxury of working with a kistler plate and cannot definitively say that they would not give the same type of feedback when working with CFO's.

The studies I have seen do not seem to give the same type of feedback as an in-shoe system, despite either systems shortcomings.

I see what I see from the FTc's and the changes in pressure patterns, CoP progression and comparitive accelerations from foot to foot. The patterns I have witnessed are the same put forth by Dr Dananberg and others who regularly use and lecture on the F-scan system.

I have to infer many of the changes for the better that I see in comparing the data I mentioned above. I would love to be much more exacting and tell you and my patients that we reduced the stress at one joint over an other by a specific percentage. In general I can tell people that just by seeing where the specific changes have taken place in the FTc's and the change in pressure patterns.

I leave it to the majority who have recently contributed to this thread to prove or disprove what I "see" using in-shoe pressure on a daily basis.

Sincerely;
Bruce

 

Eric;

When working with patients with diabetes, pressure at the ulceration is only one parameter to measure and often will not give you any indication as to whether the wound will improve or not, strictly from a decrease in pressure.

CoP progression is vitally important in working with these patients depending on where the primary neuropathic ulceration is on the foot.

FTc's are very important as well as they will give indication to whether the foot is stopping/stalling thru single limb stance and whether that is the primary contributor of the deforming forece as opposed to a prominent bone in the foot.

If pressure reduction alone were adequate to heal the wounds of patients with diabetes, we podiatrists would all be geniuses w/ or w/o any type of in-shoe pressures system!

It is as important, or even more so, to attempt to normalize the gait of patients with diabetes and chronic wounds than it is to simply offload the area of highest pressure.

Never underestimate the component of time in dealing with chronic wounds!

Bruce
PS: nice reply to Barry Block's forum the other day regarding universal health care.

 

Could someone please spell a couple of things out for me:

What's the difference between a force plate and a pressure plate?

For each device - force plate, pressure plate and inshoe system - what data is best obtained from each?

Thank you.

Rebecca

 

Hi Simon, I'd hoped someone else would be interested in this to provide the answers so I could just follow along.

I read some maths stuff the other night on straight line graphs with y=mx+c as you suggested. x and y are the coordinates, m = gradient and c is the y intercept. So I understand the gradient (steepness) of the straight line will tell you the rate of loading. The straight line at the beginning of the force-time curve (FTC) shows the rate of loading of force by the heel at heel strike. The straight line at the end of the FTC shows the rate of unloading of force from the forefoot during propulsion.

So the FTC provided earlier in the thread by Craig Payne is presumably a pretty normal representation. I'm sure the gradient of the lines at the beginning and end could look different to this - what do you see with different pathologies?

I'm sorry if this is too basic for some.

Rebecca

 

Rebecca,

I'm not sure that there is an official definition somewhere, but I'll give you my interpretation.

A force plate is a solid piece of metal with force sensors at each corner. The better force plate will have sensors in each of the planes. Thus total vertical force on the plate would be the sum of force at all of the vertical sensors. Ant post and medial to lateral force could also be measured as well as ground reaction torque. The center of pressure on the plate could be calculated by looking at the realtive amount of force on each of the vertical sesnors. However, with a force plate you would not be able to see how much force there was on an individual anatomical structure.

A pressure plate is essentially an array of tiny force plates or sensels. Each sensel produces a value of force and the pressure at that location is calulated by dividing the force applied to the sensor (sesnsel) by the cross sectional area of the sensor. The size of the sensor is critical for the pressure achieved because the same force applied to a small sensor would produce a much higher pressure than than on a large sensor. For example the force from a metatarsal head on a sensor that was just as big as the metatarsal head versus a sesnor twice that size. Center of pressure can also be calculated using a pressrue platform. However, most pressure platforms cannot give a-p and med-lat shear. But, depending on the size of senors they could give the amount of vertical force on a single metatarsal head. There is software available that can combine any number of single sensors so that you could get the force readout for large areas if you were interested. (For example the whole body) The platform is placed on the ground and is flat.

In shoe sensors are usually made from the same kind of sensors that pressure plates are made from, but they are made to fit inside of a shoe. There are different kinds of technology used to make sensors and they are all pretty cool and each has its pros and cons. The advantage of being in the shoe is that you get foot versus shoe forces instead of shoe versus ground forces. Some of the force plates are good enough that you can pick up the tread pattern on the bottom of the shoe. However, that may not help you examine the pressure under a metatarsal head.

With each system, you have to be aware of whether the technology is good enough to answer the question that you are asking.

Rebecca, I hope this gives you something

Regards,

Eric

 

As far as Eric and Rebecca’s questioning the clinical value or interpretation of the FTCs I think this is why this thread is an important discussion. I have never seen this talked about from a general clinical viewpoint. I am not sure myself how much value can be reasonably placed on this measure, in isolation I would say it is pretty meaningless other than to tell you that the gait is likely close to normal or otherwise, if steps are symmetrical, and if there is much inter-step varation for a given subject. Some will claim symmetry is important but this seems a pretty controversial idea currently.

Total contact FTCs are very sensitive but non specific because they are measuring the combined effect of all the body segments simultaneously.

Simon, I agree with your starter for 10 and would be interested to see which part(s) of the curve (image attached) you think this effect might be defined by.
Just concentrating on the first curve for now I would add this.

To be able to make any reasonable inferences, even qualitative, regarding the generalized FTC shape, it would seem to me that a synchronized frontal and saggital view of the entire lower limb would need to be examined with the force data. Knee flexion and to a lesser extent hip flexion at heel strike will have a large effect both on the amplitude and the rate loading. Try walking fast with stiff knees to experience this.

Examining FTCs for various normalized speeds which have also been normalized for body weight and time there are 2 distinct gradients for the initial vertical GRF showing on the attached image.

Does anyone feel that it is reasonable to attribute any specific variable(s) to the 3 distinct parts of the first shape of the curve? Remember that this is BAREFOOT.
I would guess that;

The initial steeper segment (A)is primarily determined by the visco elastic properties of the heel.

The remaining segements B and C determined by all of the following.

1 stride length (altering ratio of vertical to horizontal components of GRF and increasing height that COM falls and accelerates from and has to return to),

2 speed and cadence( considering inverted pendulum model , up to a certain point COM will rise faster as frequency increases),

3 duty factor (with an appropulsive gait the loading and offloading of the respective limbs will occur over a longer period and this will flatten out both ends of the curve and reduce their amplitudes because stride length will shorten).

Because all of these are interrelated I am wondering how different the normalised curves vs non-normalised ones would look for the same subject at different speeds, I have never tried this but the first part of these curves look different to what I am used to seeing.

Another thing which I don’t think has been mentioned is sample rate. An initial oscillation in the heel loading curve caused by the visco-elastic properties of the fibrofatty pad is not seen if the sample rate is set too low. I am not sure what the sample rate was for these curves but it is either lost from low sample rate or possibly filtered out.

Any comments or additions this?


DAMM . . . .. . . . . .. . I tried to upload the image but got error message stating that my quota is exceeded . . . . could anyone help out please because this post is a bit meaningless without the image :santa:

Cheers

Martin

The St. James Foot Clinic
1749 Portage Ave.
Winnipeg
Manitoba
R3J 0E6
phone [204] 837 FOOT (3668)
fax [204] 774 9918
www.winnipegfootclinic.com

 

Rebecca:

I would like to add to Eric's excellent description of force plates, pressure plates and pressure insoles.

First of all, the force plate is the only system that allows the three dimensional force vector from ground reaction force (GRF) to be analyzed. Because of this fact, the force plate is the only system that allows inverse dynamics to be used to determine moments at the joints of the lower extremity. The force plate also samples at very high rates and allows the center of pressure (CoP)_to be tracked throughout gait.

Second, the pressure mat is basically a mat with a matrix of force sensing cells embedded within it that allows the CoP to be tracked. The pressure mat also allows the GRF at discrete locations of the plantar foot to be individually analyzed which allows the foot to be divided into different areas of interest for more detailed analysis. However, the pressure mat can not detect non-vertical forces such as anterior-posterior shearing forces and medial-lateral shearing forces.

Last, the pressure insole is an insole with a matrix of force sensing cells embedded within it that allows CoP and discrete areas of the foot to be tracked, basically like a pressure mat inside of a shoe. Again, the pressure insole cannot detect anterior-posterior shearing forces and medial-lateral shearing forces.

All of these systems have their benefits and their problems. I have reviewed these technologies in an article I did a little over a year ago titled Emerging Concepts in Podiatric Biomechanics. A more recent column I was interviewed on was just published in Podiatry Today on A Closer Look At Orthotic Technologies And Modifications that discussed pressure insole/mat technology and its use in clinical practice.

Hope this helps.

 

I don't see how total contact force time curves can be that sensative as there is a fair amount of step to step variation. Any change from one condidtion to another would have to be greater than the step to step variation to be detected. Regardless, several steps need to be averaged and you cannot look at one single step verusus another single step.

Another major determinent of the initial force peak magnitude and loading rate is vertical velocity before contact. This is a behavioral variable, for example if you wanted to assess the shock aborptive properties of a heel cup you would have to control for vertical velocity just prior to impact. (A patient may dislike the sensation of hitting the ground hard in a heel cup and change their legs velocity.

.
I don't see many uses for total force time curves because of lack of specificity of expected change. If there is a change there are so many potential causes.

I am intrigued by looking at the timing of heel off as a measure of orthotic success. However, any time it is measured several steps with each condition should be averaged to ensure that the difference is not random variation. An earlier heel off might correlate with increased ankle joint power. It is certainly something that should be studied.

Regards,

Eric Fuller

 

Eric
I agree that you may be right but am not convinced yet that the variability may make meaningful comparisons impossible and am unaware that anyone has published on this subject. I suggested in my first post that this might be accomplished by using the shape factor, a single function representation of the FTC to actually be able to determine intra-subject step to step variance and pre post intervention statistical analysis. Comparing amplitude is not a problem but rate of change when non linear is.

So far nobody except Simon seems to have addressed the credibility of using the shape factor for this purpose, unfortunately my math is not up to this task and I am currently seeking out a math wiz amongst my colleagues here who may be able to help. Simon suggested simplifying the task by looking at the linear parts of the curve. I tried to attach an image to my last thread without success which would allow some further discussion on this, I will keep trying to upload this so and see what you think.

I agree that you are right regarding needing to consider need to control for prepositioning and lower limb contact velocity. If there is significant reduction in heel contact velocity I would estimate that this will manifest itself by reduced amplitude of the first curve. The effect of damping by any material underneath the heel would change the loading rate and steepest part of the curve but not the final amplitude. Provided that the amplitude was not significantly different would you not agree that the initial part of the curve could be attributed to damping since the max force applied would be the same?

I have suggested the possibility of attributing sections of the loading curve to different variables, perhaps this is not credible but so far no one has explained why this might be.

Cheers

Martin

The St. James Foot Clinic
1749 Portage Ave.
Winnipeg
Manitoba
R3J 0E6
Phone [204] 837 FOOT (3668)
Fax [204] 774 9918
www.winnipegfootclinic.com