Welcome to the Podiatry Arena forums

You are currently viewing our podiatry forum as a guest which gives you limited access to view all podiatry discussions and access our other features. By joining our free global community of Podiatrists and other interested foot health care professionals you will have access to post podiatry topics (answer and ask questions), communicate privately with other members, upload content, view attachments, receive a weekly email update of new discussions, access other special features. Registered users do not get displayed the advertisements in posted messages. Registration is fast, simple and absolutely free so please, join our global Podiatry community today!

  1. Have you considered the Clinical Biomechanics Boot Camp Online, for taking it to the next level? See here for more.
    Dismiss Notice
Dismiss Notice
Have you considered the Clinical Biomechanics Boot Camp Online, for taking it to the next level? See here for more.
Dismiss Notice
Have you liked us on Facebook to get our updates? Please do. Click here for our Facebook page.
Dismiss Notice
Do you get the weekly newsletter that Podiatry Arena sends out to update everybody? If not, click here to organise this.

Force/time curves comparison and interpretation

Discussion in 'Biomechanics, Sports and Foot orthoses' started by Mart, Feb 7, 2008.

  1. Mart

    Mart Well-Known Member


    Members do not see these Ads. Sign Up.
    Force/time curves form a complete picture of the ground reaction force underneath the foot during stance but are a fairly non specific representation of gait generally.

    I have seen case studies in which a force/time curve are compared before and after foot orthoses intervention and regarded as evidence of some useful attribute of the foot orthoses effect.

    Whilst this approach seemed to me to lack any credibility given that in these examples no indication that variance or actual amount of change that has taken place had been discussed, the idea that comparing force/time curves as an index of change seems compelling if useful interpretation can be made from this.

    I haven’t thought much about this until reading a recent paper on gait efficiency

    In this paper there is some analysis of force/time curves which involve creating a single dimensionless value for the shape of the curve.

    It occurs to me that this representation may allow an objective comparison between force/time curves in a way which I have never seen before.

    Anyone interested in this should check out page 468 fig 1

    Paper can be downloaded at

    http://www.podiatry-arena.com/podiatry-forum/showthread.php?t=7625

    post # 26


    The math is beyond my comprehension currently but using some logical deduction feel that the definitions below home in it’s essence.

    Is there anyone who can tell me if this Buckingham π theorem is what is being used here or I am barking up the wrong tree?

    If not this . . .. . .... then any pointers?

    I would like to understand this but to economize on effort to do this.

    The idea here, if I am interpreting things correctly, would allow the calculation of a single value the "shape factor" q representing the force/time curve, using this value to then consider the variance of a range of values for q over a number of steps and then determine probability of difference of q for a similar range of steps with an altered variable.

    Whilst this begs the question of how this might be useful and what interpretation could be made of this, I do have some ideas on this which I’d like to discuss further but have run out of time and wanted to develop a bit of momentum on this subject rather than submit too lengthy a post.

    Cheers

    Martin



    The compactness measure of a shape, sometimes called the shape factor, is a numerical quantity representing the degree to which the shape is compact. Various compactness measures are used. However, these measures have the following in common:


    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.


    They agree with intuitive notions of what makes a shape compact.
    A common compactness measure, called the circularity ratio, is the ratio of the area of the shape to the area of a circle (the most compact shape) having the same perimeter. That ratio is expressed mathematically as M = 4π(area) / (perimeter)2. For a circle, the ratio is one; for a square, it is π / 4; for an infinitely long and narrow shape, it is zero.

    Compactness measures can be defined for three-dimensional shapes as well, typically as functions of volume and surface area. One example of a compactness measure is sphericity Ψ. Another measure in use is (surfacearea)1.5 / (volume).[1]

    Dimensionless quantity
    In dimensional analysis, a dimensionless quantity (or more precisely, a quantity with the dimensions of 1) is a quantity without any physical units and thus a pure number. Such a number is typically defined as a product or ratio of quantities which do have units, in such a way that all the units cancel out.
    "out of every 10 apples I gather, 1 is rotten." -- the rotten-to-gathered ratio is (1 apple) / (10 apples) = 0.1 = 10%, which is a dimensionless quantity. Another more typical example in physics and engineering is the measure of plane angles with the unit of "radian". An angle measured this way is expressed as the ratio of the length of an arc lying on a circle (with its center being the vertex of the angle) swept out by the angle to the length of the radius of the circle. The ratio (length divided by length) is dimensionless.

    Dimensionless quantities are widely used in the fields of mathematics, physics, engineering, and economics but also in everyday life. Whenever one measures any physical quantity, they are measuring that physical quantity against a like dimensioned standard. Whenever one commonly measures a length with a ruler or tape measure, they are counting tick marks on the standard of length they are using, which is a dimensionless number. When they attach that dimensionless number (the number of tick marks) to the units that the standard represents, they conceptually are referring to a dimensionful quantity. A quantity Q is defined as the product of that dimensionless number n (the number of tick marks) and the unit U (the standard):

    But, ultimately, people always work with dimensionless numbers in reading measuring instruments and manipulating (changing or calculating with) even dimensionful quantities.

    In case of dimensionless quantities the unit U is a quotient of like dimensioned quantities that can be reduced to a number (kg/kg = 1, μg/g = 1e-6). Dimensionless quantities can also carry dimensionless units like % (=0.01), ppt (=1e-3), ppm (=1e-6), ppb (=1e-9).

    A dimensionless quantity has no physical unit associated with it. However, it is sometimes helpful to use the same units in both the numerator and denominator, such as kg/kg, to show the quantity being measured.

    A dimensionless proportion has the same value regardless of the measurement units used to calculate it. It has the same value whether it was calculated using the SI system of units or the imperial system of units. This doesn't hold for all dimensionless quantities; it is guaranteed to hold only for proportions.


    According to the Buckingham π-theorem of dimensional analysis, the functional dependence between a certain number (e.g., n) of variables can be reduced by the number (e.g., k) of independent dimensions occurring in those variables to give a set of p = n − k independent, dimensionless quantity. For the purposes of the experimenter, different systems which share the same description by dimensionless quantity are equivalent.

    Example
    The power consumption of a stirrer with a particular geometry is a function of the density and the viscosity of the fluid to be stirred, the size of the stirrer given by its diameter, and the speed of the stirrer. Therefore, we have n = 5 variables representing our example.

    Those n = 5 variables are built up from k = 3 dimensions which are:
    Length: L (m)
    Time: T (s)
    Mass: M (kg)



    According to the π-theorem, the n = 5 variables can be reduced by the k = 3 dimensions to form p = n − k = 5 − 3 = 2 independent dimensionless numbers which are in case of the stirrer


    Reynolds number (This is the most important dimensionless number; it describes the fluid flow regime)

    Power number (describes the stirrer and also involves the density of the fluid)

    The Buckingham π theorem is a key theorem in dimensional analysis. The theorem loosely states that if we have a physically meaningful equation involving a certain number, n, of physical variables, and these variables are expressible in terms of k independent fundamental physical quantities, then the original expression is equivalent to an equation involving a set of p = n − k dimensionless variables constructed from the original variables. This provides a method for computing sets of dimensionless parameters from the given variables, even if the form of the equation is still unknown.

    However, the choice of dimensionless parameters is not unique: Buckingham's theorem only provides a way of generating sets of dimensionless parameters, and will not choose the most 'physically meaningful'.

    More formally, the number of dimensionless terms that can be formed, p, is equal to the nullity of the dimensional matrix, and k is the rank. For the purposes of the experimenter, different systems which share the same description in terms of these dimensionless numbers are equivalent.
    In mathematical terms, if we have a physically meaningful equation such as

    where the qi are the n physical variables, and they are expressed in terms of k independent physical units, then the above equation can be restated as

    where the πi are dimensionless parameters constructed from the qi by p = n − k equations of the form

    where the exponents mi are rational numbers (they can always be taken to be integers: just raise it to a power to clear denominators).

    The use of the πi as the dimensionless parameters was introduced by Edgar Buckingham in his original 1914 paper on the subject from which the theorem draws its name.

    The Buckingham π theorem provides a method for computing sets of dimensionless parameters from the given variables, even if the form of the equation is still unknown. However, the choice of dimensionless parameters is not unique: Buckingham's theorem only provides a way of generating sets of dimensionless parameters, and will not choose the most 'physically meaningful'.

    Two systems for which these parameters coincide are called similar (as with similar triangles, they differ only in scale); they are equivalent for the purposes of the equation, and the experimentalist who wants to determine the form of the equation can choose the most convenient one.






    The St. James Foot Clinic
    1749 Portage Ave.
    Winnipeg
    Manitoba
    R3J 0E6
    Phone [204] 837 FOOT (3668)
    Fax [204] 774 9918
    www.winnipegfootclinic.com
     
    Last edited: Feb 7, 2008
  2. Martin:

    Hope that winter doesn't continue to be so long up there in Winnipeg.....from your friend in sunny Sacramento!;)
     
    Last edited: Feb 7, 2008
  3. Craig Payne

    Craig Payne Moderator

    Articles:
    8
    Martin

    Here is a pic I use in my powerpoints (courtesy Tekscan) that illustrates "some useful attribute of the foot orthoses effect" ,.... note the sooner onset of the second peak in the orthoses condition.
     

    Attached Files:

  4. CraigT

    CraigT Well-Known Member

    Craig,
    I have seen this before and it got me wondering about an aspect of this and in-shoe pressure measurement.
    When you have the Fscan insole without an orthosis, it sits flat in the shoe. When you have an orthosis, it will (more than likely) be in much closer contact with the foot- particularly the midfoot.
    Could this account for part of the apparent effect irrestpective of the kinetic changes that may occur? Would you get a similar effect with a so called sham orthosis?
    It would be much nicer to have an in-shoe system that was always in close contact with the foot... worn like a sock perhaps???
    If I am being ridiculous, I am sure that those who have a much deeper knowledge of this subject will set me straight... be nice.
     
  5. Craig Payne

    Craig Payne Moderator

    Articles:
    8
    It could be an artifact of the change in the pressure distribution, but I don't think so (we still exploring it), as I not sure how the pressure distribition necessarily changes the timing of events. If you do what is like in the picture, then jam by the first ray with a felt pad, you can see changes in the timing of the curve that would hypothetically occur. We have done a study late last year (not yet analysed) in which a number of idential cad/cam custom orthotics were given to each subject (the orthotics only differed by one small variable), so hpefully we soon get a better understanding of the shapes of that curve.
     
  6. http://www.patentstorm.us/patents/6918883.html
     
  7. Mart

    Mart Well-Known Member

    Craig(s)

    My impression of what can be reasonably inferred from these force/time curves is this:

    The red curve indicates slightly increased acceleration/deceleration of body weight and slightly advanced max acceleration of body wt during this single step

    This could be caused by a number of things including;

    1 degree hip and knee flexion at HC or both of these plus ankle plantarflexion at end of propulsion,

    2 normal variance of interstep gait

    3 systemic error of measurement system

    4 since total contact time is identical it seems reasonable to say that velocity is probably not an issue

    2 and 3 are reasons for me suggesting that looking at shape factor has merit because it would, I think eliminate these concerns.

    Because the graph represents TOTAL FORCE, I cannot see why changing the surface area of contact or timing of application of pressure due to effect of foot orthoses contact would effect this.

    The total ground reaction force will act through whatever is in contact with the sensor irrespective of its shape, the shape (from foot shoe interface) may affect body response to contact but that is different issue.

    Any problems with this because it is a very basic assumption??????

    I think that templating plantar calcaneal area and forefoot force/time curves gives us a much better picture of things and would like to look at that one we have exhausted this issue of total FTC.

    I think this is a very important thread – let’s not worry about being seen as dumbos – in my experience, having ideas revealed which are questionable is one of the best ways I hasve made progress –

    to quote from HBO’s “the wire” . . . .. . . “respect bro” :drinks

    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
     
    Last edited: Feb 7, 2008
  8. Might a simple heel lift produce the same effect shown in your F-Scan graphs, Craig?
     
  9. Mart

    Mart Well-Known Member

    I liked Simon Spooners suggestion from a different thread of creating a Pod Arena touring conference. Perhaps if that happens I could arrange for you guys to include Winnipeg in the itinery and share the suffering around a bit by hosting it in Feb ( or perhaps in June for mosquito season)

    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
     
  10. Mart

    Mart Well-Known Member


    My initial thoughts are:

    If the pelvis, hip, knee and ankle motion with and without heel lift were unchanged then we might expect the change in force/time curve to show:

    Increased amplitude of first curve because excursion of COM would be increased by longer funtional limb

    Increase of force/time integral of first curve because of longer heel contact time.


    Smaller gradient of second curve because of longer contact time of plantar calcaneal area causing slower offloading of force from rearfoot onto forefoot.

    Because the lower limb function will probably change in response to heel ramp, and that change would be very subject specific, any changes in force/time curve shape would be impossible to explain without looking at entire lower limb function.

    Separating total force into separate plantar calcaneal area and forefoot force/time curves would give us a much more interesting impression because it would reveal more information about the respective plantar calcaneal area and forefoot loading and unloading velocities and how timing changes.

    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
     
    Last edited: Feb 7, 2008
  11. Asher

    Asher Well-Known Member

    I think I should know this:

    In comparing two steps from the same person (as in Craig's Tekscan image), is the area under the force/time curve going to be exactly the same?

    If you compare a force/time curve for a walking step and a running step, is there still going to be the same area under the curve inspite of the curve being a different shape?

    Rebecca
     
  12. Asher

    Asher Well-Known Member

    I found the concept of duty factor interesting. Duty factor describes how long a foot is in ground contact for per stride. In walking it is about 0.6 (spends more time on the ground than in the air), in running it is about 0.35 (spends more time in the air than on the ground).

    You don't want the duty factor to be too high as it indicates the foot is in contact with the ground for too long which gives it less time in swing, which means muscles have to work harder to get the leg through swing and on the ground again for the next step.

    It seems this is a measure of delayed heel lift which I look for on video gait analysis but have no idea at what point it becomes labelled as delayed.

    At the same time, you don't want the duty factor to be too low because it means GRF is very high.

    Is duty factor measured simply by timing one stride and determining the fraction of stance time / stride time? If so, sounds like it would be relatively easy to define delayed heel lift using Tekscan gear?

    Rebecca
     
  13. Mart

    Mart Well-Known Member

    Not neccessarily, it depends how the body (primarily the lower limb) deals with the motion of the body COM which is mostly sag plane motion at hip, knee and ankle and a little bit of pelvis.

    The peaks are determined by amplitude of COM excursion and the shape of the curve by the rate of change, this is completely different for walk and run because there is a shift from primarily stiff knee biphasic "pendulum" excursion to monophasic "spring" knee much bouncier excursion of COM and no double support.

    Best intro text for me was Chris Kirtly's Clinical Gait Analysis which pops up in the Amazon box to the right from time to time. If the above dosent make sense go buy it and read it . . . . . . .. . twice.

    BTW any chance of getting your math buddy to look over my initial post and see if he rolls his eyes or offers to become the mailbase maths guru?"


    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


    .
     
    Last edited: Feb 8, 2008
  14. Mart

    Mart Well-Known Member


    Here’s my understanding;


    Delayed heel lift may or may not increase duty factor (more likely that it would).

    The issue is moslty to do with propulsion of swing limb which is most efficient with smaller double support time (allows ankle plantar flexors to contribute more concentrically, in appropulsive gait they don’t really contribute much?) but is limited by the stress caused by shorter contact period and maximum power available to muscles inputing into swing limb (mostly ankle plantar flexors and hip flexors)

    I would estimate that in real world for a given subject at given instant in time for walking gait;

    that higher duty factor would equate to shorter stride length, lower velocity, reduced ankle dorsiflexion during stance, less power needed for eccentric contraction of ankle plantarflexors, little opportunity for concentric contraction of ankle plantarflexors, lower amplitude of FTC humps and greater demand on hip flexors to move swing limb forward.


    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
     
    Last edited: Feb 8, 2008
  15. Asher

    Asher Well-Known Member

    Hang on, a stride is one heel contact to the ipsilateral (same foot) heel contact.

    To have a duty factor of one, your foot wouldn't leave the ground ie: glide it over the floor.




    Depends on speed. Sorry Martin, what is the point you are making?

    Agreed

    What do you think about measuring duty factor, do you think it could be helpful?

    Rebecca
     
  16. Asher

    Asher Well-Known Member

    Buggar, that would have been convenient. :rolleyes:

    I have emailed it to her, will let you know.

    Rebecca
     
  17. Mart

    Mart Well-Known Member

    Hi Rebecca


    Sorry . . . . .. I just made a corection to what I wrote in this post which was wrong, and you are of course quite . . .. right duty factor of 1 .....duhh

    note to self

    "go home at end of long day at office and have supper before contributing to nonesense to mailbase"


    anyhow well spotted . . . .. . . glad someone is reading this stuff

    not sure what you are refering to on speed though

    please re my post again which I think now make sense


    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
     
  18. Mart

    Mart Well-Known Member


    I think that it was essential for the purposes of the model, clinically we are aware of this intuitively from just looking at the kinematics.

    Because it's value is speed related though that I am not sure if it would be clincally meaningful in its own right.

    BTW thanks for emailing to your math pal - sorry about the gender assumption/bias . . . geeze 2 blunders in one post . . . :eek: . . . . I need to go have a beer.


    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
     
  19. Asher

    Asher Well-Known Member

    I agree that we see it when doing a video gait analysis. But there is not a definition for 'delayed heel lift' and I just thought that duty factor could provide a definition - I could be wrong.

    But for walking, there is a fairly pinpoint speed which is most economical (walking economy curve) and is our self-selected walking speed.

    Rebecca
     
  20. Asher

    Asher Well-Known Member

    Does anyone have the following paper available?

    Alexander RM, Jayes AS. Fourier analysis of forces exerted in walking and running.
    J Biomech. 1980;13(4):383-90.

    Thanks

    Rebecca
     
  21. Bruce Williams

    Bruce Williams Well-Known Member

    Kevin, Martin, Craig's, Simon;

    A heel lift might or might not effect the curves in such a way.

    Craig is correct that when the orthotic device conforms more to the foot, the medial foot pressures often will increase or at the very least register when they did not before the use of a device.

    Craig P. is also correct that this does not necessarily effect the kinematics of the gait curve.

    Martin has made excellent points as well on variability, etc.

    Ultimately it comes down to many differing factors to register improvement utilizing the gait curve. Sometimes it is ridiculously easy, and sometimes very difficult.

    Limb lenth plays a huge roll. unblocking stoppages of the FF pivot is important..thru use of cutouts, FF wedges / posts, cluffy wedges, etc.
    Finally getting the AJ to move makes a huge difference as well.

    Cheers! I'm off to the first annual AAPSM Athletic Shoe Reccomendation Committee retreat in warm and sunny Chicago!
    Bruce
     
  22. Mart

    Mart Well-Known Member

    Hi Bruce

    I was hoping that you would add your voice to this topic.

    I am curious about your reasoning for sseing change in shape of curve according to conformity.

    If you walk over a Kistler plate it contains one single sensor for vertical ground reaction force. To my mind the force/time cirve (FTC) should look the same regardless of whether measured with in shoe matrix senor or Kislter plate.

    Since we are looking at total force not pressure, provided that all the sensors that have contact with the foot are functioning correctly conformity should not matter. With a FO normaly there will be increased surface are of conatct so pressures may be reduced or increased in certain area but total force and its reflection of acceleration of body mass would not be effected becasue of this effect.

    I am skeptical but curious about your emphasis on LLD and would love to expand on this particarly with ref to FTC interpretation.

    perhaps we could start with a theoretical overview of how this might work, and then move on to some case studies?

    I'll post my thoughts later when I get a moment


    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
     
    Last edited: Feb 8, 2008
  23. Bruce Williams

    Bruce Williams Well-Known Member

    Martin;

    I don't think I said that the FTc's would necessarily change due to conformity of the orthotic device. The pressure picture will usually change, though not always, b/c of increased conformity of the medial arch and pushing the pressure sensor into that area.

    I do not have much experience with kistler plates. I've seen the data from the studies, the FTc's, etc. I often find that the data collected does not necessarily jibe with what I usually get when utilizing in-shoe pressure sensors. I assume that is because of the shoe and its effects on the foot function. Barefoot data is interesting, but in general I'm much more intriqued by how the foot functions in a shoe w/ and w/o a CFO (custom foot orthosis).

    Regarding LLD and in-shoe pressure data. Much work has been reviewed by some terrific articles in Gait and Posture, from Leg length discrepancy – an experimental study of compensatory changes in three dimensions using gait analysis. Walsh, Connolly, Jenkison, O’Brien. Gait and Posture 12 (2000) 156-161.
    That covers 3D analysis and it corresponds to what I see in-shoe most of the time.

    Basically I see LLD working in a couple of different ways. 1) if the MTJ is stable on the short limb or has high DFiion stiffness with the metatarsals, then the heel will lift off the ground early and will accelerate to the mpj's faster than the opposite long sided limb which is usually less compliant. Usually you will see high sub 1st mpj pressures, or higher than on the long sided limb.
    2) false positives will show up on the long sided limb when the MTJ is overly compliant adn you will see early heel lift adn then a stoppage of the CoP. movement in teh midfoot area and often low pressure under the 1st mpj. This is usually associated wtih early knee flexionon the long sided limb to compensate for the short sided limb.
    3) finally, you can have LLD w/ a tranverse abduction rotation of the foot that can often skew the pressure readings as well. These feet usually will have a more compliant MTJ, though not always. The heel lift will often be somewhat delayed as they roll from the lateral foot for a prolonged period of time.

    That's all I can add for now. Let me know your thoughts as I am always willing to learn something new.
    Cheers.
    Bruce:good:
     
  24. Craig Payne

    Craig Payne Moderator

    Articles:
    8
    People with diabetes do not get ulcers walking around barefoot; athletes do not get overuse injuires running around barefoot ..... they get their problems wearing footwear, so who cares about what happens barefoot over a force plate or pressure platform?
     
  25. efuller

    efuller MVP

    I believe that a Kistler plate has four vertical sensors, one at each corner. If it had one it could not calculate center of pressure. Center of pressure would be calcualated by taking a weighted average of the four sensors.

    I agree that in shoe force total should be very close to the shoe versus force plate force total. The difference should be roughly the weight of the shoe. I was at the 2nd EMED users conference in 1991 and someone did a presentation on how the in shoe force was 2/3rds of the shoe versus plate force. Peter cavanagh pointed out that this is more likely a problem with in shoe sensor that a true difference between conditions. I'm sure they've improved since then. There are calibration issues with any of these devices. I've used the FSCAN about 10 years ago and they had something called an extended calibration routine that took the patient's body weight and made sure that the force time curve had magnitudes that would make sense in relation to the body weight. So, the data may be "cleaned up" by the software so that you may not be looking at what was really measured. The calibration method is really important. Has anyone compared the measurement output with calibration with a bladder versus calibration with the patient standing on the sensor? Or calibration with the sensor on the orthotic or off of the orthotic. This should be published before we accept small differences in the force time curve as being significant.


    Cheers,
    Eric Fuller
     
  26. Mart

    Mart Well-Known Member

    Thanks Eric, I stand corrected re number of sensors on a Kistler Plate but you have I think understood the point I was trying to make.

    Craig perhaps I did not explain my point clearly - either that or I am completely out to lunch in my understanding. I'll try and do a quick experiment tomorrow to clarify what I meant which is consistent with Erics 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
     
  27. Mart

    Mart Well-Known Member

    I have had the same concerns regarding accuracy, just got a bladder system for my FMat and inshoe FScan sensors and am hoping to set it up this weekend.

    Although this will compensate for non linearity of the sensor (the jargon being "Equilibration" ) it is not true calibration

    here' the help file info on this


    "Each sensor sensel is somewhat unique - the process for creating sensors results in some variation from sensel to sensel. In addition, as a sensor is used, some areas may become less responsive than others. The purpose of the equilibration process is to electronically compensate for these slight differences. "Equilibration" is accomplished by applying a highly uniform pressure across all individual sensing elements. With a uniform pressure applied, each sensel within the sensor should produce the same output. When this is not the case, the software determines a unique scale factor for that sensel to compensate for the slight variation."



    Tekscan recommends calibration using body weight also suggesting that it is done with best approximation to test conditions to achieve optimal accuracy, several different options are available with the research sofware which I am using.


    I would anticipate that if the sensors are accurate and actually calibrated as well as equilibriated in the bladder then test measurements taken inside the bladder should be as expected. I want to find out if this is true.

    If I get time I'll post the pre and post calibration data for the bladder experiment and also for the test I am planning to post for CraigP's comments

    Craig I am still purplexed at your apparent ambivalence to the value of barefoot measurement :confused: .

    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
     
  28. Mart

    Mart Well-Known Member

    Craig

    Thinking about this a bit more:

    Surely pressure mat systems are important compared to in shoe systems for 2 fundamental reasons and situations, and they are

    Comparison with of subject with normative values. It is impossible to create normative values for in shoe measurement because of immeasurable effects of influence of foot-wear.

    Ans as you already mentioned - sensor artifacts.

    I was going to walk over my FMat shod and compare total FTC to in shoe measurement with same shoe, foot and speed to check that I am not confused but this has already been done (not by a psychologist I might add :eek:) much more accurately than I could, FTC (attached JPEG image from paper below).

    Generalizability of in-shoe peak pressure measures using the F-scan system.
    Mueller MJ, Strube MJ.

    Program of Physical Therapy, Washington University School of Medicine, St Louis, Missouri, USA.

    In-shoe pressure analysis can be useful in the management of a variety of foot and ankle problems, but guidelines are needed to determine the practical limitations of the measures. The primary purpose of this study was to determine the reliability of peak plantar pressures taken with the F-Scan system over multiple steps, sensors, and days, and using a force platform for additional calibration. Data were collected on 10 healthy subjects as they walked across a force platform for a minimum of three trials on four separate sessions that were 1 week apart. Using a mean of three steps with a single sensor on 1 day, generalizability coefficients were 0.75 for manufacturer calibration and 0.82 with force platform calibration; reliability coefficients for absolute decisions (index of dependability) were 0.60 for manufacturer calibration and 0.76 with force platform calibration. Force measures from the F-Scan and force platform were highly correlated (r = 0.93), but the absolute difference between the measures varied between sensors and over time.

    RELEVANCE:--Peak pressure measures taken from a mean of three steps with procedures outlined by the F-Scan manufacturer provide pressure values that demonstrate adequate reliability for clinical and research purposes when rank ordering of measures is indicated. If measures are to be used for making absolute decisions (e.g., identifying a threshold for injury), calibration from a stable source such as a force platform is recommended.

    Whilst the entire foot FTC has limited value compared to other data representation available from in shoe systems (ie regionalized values) surely it has an important place especially considering the problems you mention identifying artifacts. Wouldn’t simultaneous measurement of in shoe compared to Force plate data as was done in this study help answer your question regarding these artifacts?

    I would estimate that in shoe sensor artifacts are likely to result from large contour gradients which are mostly found around the plantar calcaneal area and margins of cutway window to first metatarso-phalangeal joint bending marginal sensors.

    I had suggested in another post on FHL that placing an in shoe sensor under a foot orthoses and comparing this to on top of foot orthoses might shed some light on this (interpreting only total sensor FTC ).

    I have never seen this reported and am curious about why this has not apparently been considered (other than its limitation being total force and not giving useful pressure data)

    As a crude indicator I have emulated the effects of marginal sensors by comparing the identical step data (from an FMat) from someone with abnormal “second hump” which I suspect is FHL related.

    Total sensor measurement is compared to a mask determined by tracing a new perimeter obtained by raising the lowest included force threshold.
    As might be expected the curve is different but has some similar qualities.

    My point here is to suggest that it is unlikely that there will NOT be marginal artifacts.

    Therefore even more reason to measure, at least in part, the interface between foot orthoses and shoe or shoe and ground rather than simply foot and foot orthoses.

    I hope this explains my point more clearly

    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
     

    Attached Files:

    Last edited: Feb 10, 2008
  29. Bruce Williams

    Bruce Williams Well-Known Member

    Martin;
    from what I see of your Fmat data I would not disagree that there is a sign of FnHl. I would add that there seems to be the potential for some AJ loss of pivot or a delayed late calcaneal unweighting as well.

    We could examine this more if you would now outline the rearfoot and Forefoot areas and generate FtC's. This will give you more information for determining what may be going on.

    Further, as I'm sure you know but others may not, the pressure profiles from a mat are quite different from Fscan in-shoe pressure data. There is much more indication of transverse plane transition, or abduction, when utilizing and fmat, that cannot be determined to the same level when utilitinz in-shoe pressure. This does not generally affect the FtC's , but could potentially.

    Talk to you soon.

    bruce
     
  30. Mart

    Mart Well-Known Member

    Bruce

    What I find interesting about your comment is how its illustrates the limitation of just looking at the total foot FTC.

    CC currently is exercise induced LF metatarsalgia, (only with fast walk resolution quickly with rest) No pain with competitive XC skiing.

    I did a pretty thorough gait exam of this person because of difficulties being sure of cause(s) of pain and looking to optimize FO since I suspect mechanical cause.

    He has the bilaterally the highest lunge value I have recorded (approx 50 degs). But he achieves this with exaggerated MTJ dflexion and complete ILA collapse.

    My impression is of medial devn of STJA.

    Good ROM at 1st MPJ but locks with manual DF of 1st ray bilat

    If you look at a saggital video he appears to have a delayed heel rise, but a masked FTC exam comparing plantar heel and FF showed very rapid and early offloading of heel, (assumption is that on video it is difficult to see decompression of fibrofatty pad at HR.

    He has moderate abductory twist at HR.

    His FTC is abnormal for RF only but symptoms LF only.

    He has a hx of mechanical PHP (compression and tensile perpetuated) which resolved completely with custom FO.

    He has below normal ff PP and FTIs MTHS at most MTHs at normal walking speed.

    No signs of abnormal force at midfoot. (I suspected calibration error but checking for body wt this seems unlikely)

    No skin lesions or evidence of PVD.

    Slight reproducible tinels signs LF only at MM results in tingling @ PMA

    Left side excessive internal rotation seen by patella alignment thro stance Right is normal

    LS a rather unusual knee alignment which places femur slightly medial to tibia and noticeably asymmetric.

    No evidence of significant LLD which I can detect.

    US exam of tarsal tunnel negative for space filling lesion or tibial nerve thickening

    My working hypothesis is Tibial nerve compression related to abductory twist/ and increased pronatory velocity at higher walking speeds.

    I will try very medially positioned 16 deg med RF post with high flanges to attempt reduction of pronatory stress at HC thro MS.

    Any comments or request for more clinical observations to explore this further welcome


    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
     
  31. Mart

    Mart Well-Known Member

    Craig(s)

    Considering this again it seems most likely, (assuming no significant hip, knee, ankle changes), that the earlier heel offloading and earlier “2nd hump” are caused by earlier contact with the foot against the supporting surface anterior to the heel (ie arch of FO) than before (no FO).

    This would slightly shift onset of mid-stance forward and I would estimate degree of effect (time shift and reversal of COM acceleration) be related to contour and stiffness of FO material.

    I would imagine that this is how you would see this and is I think what Craig T was thinking about it along same lines, am I correct?

    If this is true I am curious about why those considering this post think this might be therapeutically advantageous?

    Does anyone think that if the sensor was positioned between the FO and the shoe rather than the foot and the FO the FTC would look substantially different?



    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
     
  32. Bruce Williams

    Bruce Williams Well-Known Member

    Martin;

    though I am very guilty of this myself on a regular occasion, I ask that you at least once identify the abbreviations that you use. You lost me after the 2-3rd sentences.

    Regarding your statement that the patient has the best lunge test you've ever registered, but that he accomplishes this thru MTJ DFion, this seems to prove my observations of the Gait Curve you showed.

    I think that it is a common misconception that AJE is not associated with MTJ DFion compensation. I could be not understanding your post though.

    YOu state that he has a fast offloading of the heel. That is not uncommon with the foot I think you are describing. The point is not so much whether the heel offloads, but does teh Center of Pressure icon continue to move forward or does it slow or stop drammatically prior to continuing forward to the Forefoot.

    What I described above would explain the flat erratic portions of the gait curve you showed before. There is a stoppage of forward progression in the midfoot. The abducted foot postion on the left, the internal position of the talus will all lead to this.

    Regardless of whether your Physical Exam (PE) determines a LLD, you will more than likely have to treat the patietn with some form of heel lift one side or the other to help to equalize his acceleratins and get him moving or rolling or pivoting as well as possible on this left side.

    Bruce
     
  33. Bruce Williams

    Bruce Williams Well-Known Member

    Oh it will look substantially different all right! essentially, uselessly different.

    Keep in mind we want to know what the foot is doing because of the orthotic, not what the shoe is doing because of the orthotic.

    If this worked well, then there would be no use for in-shoe pressure as we could just walk over force plates in our shoes adn get perfectly useful information. It does not work that way at all.

    Also, depending on what type of FO you utilize, this will make huge pressure points on the sensor when used the way you suggested above.

    Bruce
     
  34. Mart

    Mart Well-Known Member

    Bruce

    Sorry about abbrevs I did this in a bit of a hurry.

    I think this might make an interesting case study and when I get time I'll post a proper description and also analysis of forefoot to rearfoot FTCs which I typicaly use (modified Norm Murphy's so called 3 box idea) and some ideas I am trying figure out to evalauate what you mention of the speed of COP in the forefoot.

    I am really curious to try an understand implications of these 3 box curves and how to interpret them accurately without making false assumptions.


    what I am really interested in hearing about from this thread so far is what people think about the idea of comparing the FTC of the same foot/ foot orthoses/ shoe / inshoe sensor both between foot and foot orthoses and foot orthoses and shoe as a way of identifying presence of sensor artifacts which may be caused by bending of sensor around steep contours.



    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
     
  35. I don't think that this will achieve your objective. That is, comparing the sensor data on top of the orthoses with data from the sensor positioned beneath the orthoses will not prove anything about sensor artifact.

    In terms of measuring force/ time, I'm interested in the notion that it is the pressure between the feet and the orthoses that is important as oppose to the force exerted by the foot + orthoses + shoe against ground and vice versa. If we are interested in the force/ time curve why is this better data than that obtained from a force plate e.g. Kistler, that can provide 3D analysis of force/ time?
     
  36. Martin,
    I've not had much chance to contribute to this thread, but if we go back to your first posting here regarding the "difficult math", if the mathematics employed to describe the whole force/ time curve is beyond your comprehension why not break the force time curve down into it's components? If you are keen to engage with research, and you have intimated that this is the case, then the big straight lines at the beginning (and the end) look pretty simple (y= mx +c), and from this you could glean some important information, perhaps look at the first "hump" and treat it as a quadratic/ cubic (depending how long during stance your interested in). Hint: think about viscoelasticity, rate of loading and the effects orthoses variables might have on this ;) You could do this on your daughters calculator. Have fun, look forward to reading you in print.

    K.I.S.S. (keep it simple stupid)
     
  37. Mart

    Mart Well-Known Member

    Bruce

    Thanks for your reply but I think I did not explain my idea to you properly.

    I know that the pressure distibution will look entirely different depending on sensor location.

    What I am thinking about is the TOTAL FOOT FORCE time curve which I cannot see should look significantly different except for sensor artifacts.

    Perhaps I am wrong and before I spend any time exploring this, I thought it was a worthwhile "thought experiment" partly so that I dont waste time but also because it encourages (at least for me) a deeper understanding of what we are looking at.

    do you think the FTC curves will look different and if so why?

    I notice Simon has made some comments which I want to consider a bit more before replying

    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
     
  38. Mart

    Mart Well-Known Member

    Simon

    I agree with you about the relative value of pressure change vs total FTC. Likewise Kister will be much more accurate (most studies calibrate their inshoe systems with a force plate).


    Seems like from Bruce and your responces that I have not made this understandable, hope the following makes better sence.

    My idea is that if, as CraigP suggested, significant sensor atifacts occur with inshoe measurement, and if these, as I suspect, can be attributed mostly to "crinkling" and "edge effect" then those contibutions might be discovered by removing them (sensor at FO / shoe interface) and comparing total foot FTC from this postition to FO / foot interface (not considering spatial changes in pressure.

    What do you think of is this given my assumption that visco elastic effect of FO might be negligable or predictable on shape of total foot FTC?

    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
     
  39. efuller

    efuller MVP

    Martin,
    Your question is does a sensor output change if it significantly bent? To answer this question you should remove the sensor from the foot and shoe entirely. Start with the sensor on a flat surface and put a known load on it and see what it measures. Then take surfaces with smaller and smaller radius of curvature and load with a known weight and measure again. This would be easier if you had a bladder that could apply a known pressure to the sensor. Once you find a radius of curvature that degrades the signal you can stop. EMED used to publish the radius of curvature which was too low for the sensor to properly work.

    There is also the issue of folding. Is the sensor (or sensel) destroyed when it is folded (crinkled). Can it be recalibrated to then produce reliable data?

    Once you know the radius of curvature that you start seeing output being altered you will have to compare that to the radius of curvature or the orthotics that you are going to put the sensors on.

    I don't see any way that you could figure this out using people and orthotics, because you need to know what the correct output would be and compare that to the actual output. The forces throughout gait are too variable to "know" what the correct output should be.

    Regards,

    Eric Fuller
     
  40. I agree with Eric's methodology to resolve this.

    I don't think you can judge this from the force/ time curve. I was suggesting earlier that you may look at the influence of foot orthoses on loading rates from the force/ time curve by segmenting the curve and calculating the gradients of feet with and without or with different types of orthoses.

    Let's go back a step or two: What data can be obtained from the force/ time curve?
     
Loading...

Share This Page