Postural Balance - Science topic

Sorry for the late comment.

But I have been using bestest, and this could be useful for you.

Cheers.

Sorry for the late comment. But I was reading the conversation. Does any of you know where to find normalized posturographic values- in some postures? Thank you

Thnak you Dariusz...

and what you thinking on longitudinal monitoring of large group of participates via mobile app and using of data for science?

Dragan

I am sure that my response, in that it makes a point, also inherently carries with it the potential for the reader to choose to take offense. Nothing is shared in the spirit of arrogance or petulance, but with the certainty of one who has smacked my palm to my own forehead over my own ignorance.

1. Newton never said "equal and opposite and identical". The ankle possesses DRL, or divergent reactive leverage. In the direction of gravity, it manifests as a class 2 lever. In the reactive direction known in physics by the term "normal force" it manifests simultaneously as a class 1 lever. Both must be modeled accurately in motion analysis.

Question: are these truths not self-evident? If so, then if any research draws any conclusion based upon modeling that excludes either type of leverage, is not that research merely validated by flawed premises, and thus its soundness cannot be assumed?

2. Pain leads to biomechanical changes. This was proved by Pohl 2015 trying to replicate an earlier gait analysis of lesser glute weakening and inquiring about compensatory trunk lean. While kinematic changes were noted in the earlier study, none were seen in Pohl. He told me that there was tremendous resistance to him publishing due to... well... due to people not knowing everything about everything. You see, the earlier study used painful hypersaline injections to weaken the lesser glutes, but Pohl used a painless nerve block. While the other study showed changes, there was no kinematic change in Pohl's subjects. Now, before everyone gets their collective knickers in a twist, understand that Jacquie Perry got it wrong. There isn't one gait paradigm; there are two. Just like the new Chevy truck with a V8 that can also run as a 4 cylinder depending on terrain, the human body has TDGC, or terrain-dependent gait configurations. Only on level terrain do we fully extend the hip and knee in normal gait. Both gastroc and TFL require full extension to be allowed to participate. Where was Pohl's study? In a gait lab on level ground. Have the subjects walk down a hill and things change drastically.

3. Asymmetry leads to biomechanical asymmetry. Whether LLD or muscle, there will be an impact. Give it long enough and Davis' Law can mess with you and cause pain. Bell 2013 looked at short transfemoral amputees and long transfemoral amputees. Both walked with the same gait efficiency, but the short limbs had more trunk lean. While Bell stuck with "compensatory motions are inefficient", her data proves that the truth is "compensatory motions effectively preserve gait efficiency". But here's the point... at what cost? Does the greater trunk lean of the shorter TF amputee lead to back pain sooner? We don't know because Bell had the wrong assumptions.

Gentlemen, one cannot use a linear process in a situation with reactive elements. And everything with gait reacts. We must engage in abductive reasoning prior to either inductive or deductive reasoning. In the words of Carlo Rovelli, the theoretical physicist, "Before measurements, calculations, and meticulous deductions, science is above all about visions. Science begins with a vision. Scientific thought is fed by the capacity to "see" things differently than they have previously been seen."

What are you gentlemen missing? Perhaps that biomechanics itself is limited in its vision. Page 491 of Neumann's kinesiology textbook tells us that with the hip, regardless of whether it is the pelvis or the femur moving, we see the hip osteokinematics in anatomical position. Perfect! You see the pelvis as a class 1 lever. Except... if you have the capacity for vision, you might turn your head sideways 90 degrees and see the femur as a class 3 lever.

And folks... if you've never done that, you don't know what you're missing. You see, the hip is not the class 1 lever espoused by the AAOS and the ASB. It is a UCL, or a unified compound lever, which is multiple levers that are simultaneously supported by a ball fulcrum. For is not the femoral head a circle? Why must you limit your vision to only vertical when the very model violates Archimedes' Law of the Lever (glutes attach to femur, which is the fulcrum-thus the distance value is zero. The hip is mathematically irrational allowing it to include only vertical forces).

Or perhaps it's that biomechanics only utilizes a single simple machine when there are six from which to choose. For the man whose only tool is a hammer, the whole world is a nail. For the researcher whose only tool is a lever... but there is also a wedge, a pulley, an inclined plane, a wheel and axle, and a screw. If your only tool is a lever, you must protect it, you must limit it, you must isolate it. This is the case with "closed chain" movement. Dennis Shavelson , you will find this JOSPT editorial fascinating Closed chain was only supposed to refer to structures, not movement. By only having levers, we create our own errors. In a sense, closed chain allows us to make energy simply stop, in clear violation of the second law of thermodynamics. When we allow abductive reasoning and the vision of the dreamers, we realize that vastus lateralis acts as a heterogeneous wedge to counter the lateral thrust of the femoral lever at the hip.

Trust me... I know... because during transfemoral amputation, the IT band which holds everything in place is cut. Suddenly, the self-neutralizing horizontal forces acting at the hip manifest. And the answer of the AAOS was to remove the IT band and weaken the hip abductors in the name of "balance". Forever my patients have struggled missing 30% of hip power (Ryser 1988)".

Is there more? Yes... but I think I've said enough for today. With those facepalm epiphanies, I've come to be far more aware of unnecessary patient suffering.

My apologies if this appears disjointed. The flawed assumptions in biomechanics are so numerous that it makes it difficult to only address one. Suffice it to say that if your initial approach includes a fatal flaw, everything thereafter is vulnerable as well. While the actions of biomechanics are organisms... unless you are pushing against someone's foot instead of against the ground... those reactions are mechanisms and very different in nature than the organism. And the reactive nature of mechanisms, while foreign to most in healthcare, is all we know in prosthetics.

Hi Dragan, E. Jaspers and E. Butler have both created an upper limb model to be applied in Vicon. The protocol developed by E. Jaspers has an open-source software toolbox (with matlab) which can be found online.

Mastication of chewing gum improved the postural stability during upright standing in some studies. Explanation may lay in activation of mastication muscles, condition of temporomandibular joint and trigeminal proprioception. I suggest this study to confirm the opinion:

Look at cardiology paperworks related to simpato-vagal balance in heart rate variability (studies on RR and RT heart rate variability). You should be able to have some insight in the healthy volunteers during standup/tilting position and effect on blood pressure.

Today data collection can easily be achieved recording a dynamic ECG of the subject and observe the level of Heart rate variability. Hope it helps

ONe leg standing and Monica van Eijk has in here thesis CRAMPS this test little correct esepcially in time. Rikli and Jones have test of 7 itens that all are possible by dementia p[atients and give an view what this patient can compare with even old individuals with no dementia

TUG test is possible and the BBS is possible but often to ngreat for dementia patient to do in one time

I further try to investigated the perception and the only test that was good possible was the test for the gnostic sense with an tuning fork.

succes

jan van de Rakt

I think it all comes down to matching your outcome measure to your research question. For instance, if you want to look at variability of a dataset, perhaps the standard deviation of the COP or even a nonlinear approach such as approximate entropy to look at the randomness within a sample.

Another example is using fractal dimension, which looks at the 'predictability' of a given path, where one would assume a more complex path represents either an altered postural control strategy, or perhaps dysfunction of the sensorimotor system.

I suggest reading this article by Prieto et al, 1996, regarding various COP outcomes, and it includes the calculations for each: http://www.ncbi.nlm.nih.gov/pubmed/9214811

Within my lab, we use time-to-boundary, which is a derived signal that looks at the direction and velocity of the COP signal with respect to the borders of the foot. See Hertel et al, 2006: http://www.ncbi.nlm.nih.gov/pubmed/16760569 There are other outcomes like this, and this would represent the amount of time it would take for an individual's COP to pass outside the borders of their foot (base of support), thus the assumption is made they would lose balance or be in a compromised position.

The 95% confidence ellipse is also a popular outcome, representing the locations of the COP, or another measure of COP area, if you want to look at the area of COP displacement.

Ultimately, my first point will help you best. Figure out which outcome will tell you most about your population or your balance task, and then use it. 

I hope this helps,

Hi

The therapeutic approach to this child could be much different among therapists of different schools of clinical thought. Furthermore, the fact that each child needs individualized management allows only for some suggestions about your approach rather than specific exercises. Personally, I would suggest you not to focus on the child’s muscle tone. I would choose a more functional approach. For example, if the aim is to improve his/her walking ability, I would provide many repetitions of this task by continuously changing the personal and environmental demands. I would practice gait in different conditions each time without focusing much on the quality of the movement. If you would like to additionally intervene at impairment level, I would suggest you to analyse the movement (the kinematics of her/his gait) in order to make specific hypotheses for the underlying impairments (e.g. muscle strength), to test clinically these hypotheses and then if these hypotheses are confirmed to provide specific interventions for these impairments. I think that such a task analysis and clinical reasoning could reveal function-related impairments for interventions.

Hope this helps

Zach

Ilker,

I believe you are asking whether APAs are present during more functional locomotor activities. Compared to unobstructed, overground locomotion, we know that the individuals alter a variety of gait parameters and the trajectory of the swing limb as they approach the obstacle (see some of the work from Shirley Rietdyk, Aftab Patla, and others). Changes to posture necessarily occur prior to the transition from dual to single limb stance and serve to maintain the center-of-gravity within the base-of-support of the stance limb and rotate the body over the stance ankle to maintain forward momentum. If APAs are considered compensatory muscle activity (e.g., co-contraction) or limb movement occurring prior to the execution of a focal or goal-directed behavior, then I would presume that you would observe altered muscle activation patterns of the limbs and/or axial segments in the steps preceding and following the obstacle. In order to get at your question, you could perhaps compare the kinematic, kinetic, and electromyographic variables (whichever ones suit your question) during unobstructed and obstructed walking to examine movement alterations that occur in response to this environmental challenge. However, in my opinion, dichotomous perspectives of motor control that treat "postural" and "goal-directed" movements as fundamentally different are unnecessary. Movement dynamics are consistently altered with respect to environmental manipulations in a manner that supports the completion of higher order behavioral goals.

Best of luck with your work!

Josh

Marcus,

Calculating the velocity is relatively straight forward in Excel, but you need to know a few things first: your sample frequency and the units of measurement. I'm not sure why you have z as an output signal as that is traditionally the vertical dimension, of which the COP should not be changing. The x and y coordinates are more often used, to represent AP and ML position of the COP depending on the orientation of the participant.

Let's say you collected a 10 second balance trial at 100 Hz. Column A is the X coordinate of the COP and Column B is the Y coordinate, representing AP and ML displacement respectively. My force platform's output is in cm, so I may get coordinates such as:

       A         B

1    1.15, 0.30

2    1.20, 0.34

       ...... , ........

In column C, starting at C2, just calculate the velocity between each data point as follows: C2 = (abs(a2-a1)/0.01) or (abs (datapoint _x+1 - datatpoint _x)/(1 / sample frequency) ) and you would get your ouput in cm/s, which can easily be converted into m/s if you want. Then get your average of column C and you should have your average velocity for the trial in each coordinate plane.

The 95% area is possible in excel, although it uses some methods such as singular value decomposition, which I am not as familiar with in excel (it is much easier in matlab). So I am sorry I am unable to help with that measure.

The Nyquist sampling theorem dictates your sampling frequency. You first must determine the highest frequency intrinsic to the process you are trying to record. The Nyquist theorem states that your sampling frequency must be at least twice that of the highest intrinsic frequency. If you don't, aliasing will occur. Often, the ideal sampling frequency may be more like ten times that of the highest intrinsic frequency. 

Dear Javier,

I've found some papers which deal directly with this subject.

I hope this helps.

In the early 1980s, I had the opportunity to work with Dr. Anne Shumway-Cook to develop an early postural sway biofeedback device.  In the course of our early clinical testing, one of the PTs on the team became pregnant, which posed the question, what happens to postural sway during pregnancy.  Because our device was an early prototype, we recruited another female therapist to act as a normal, and measured postural sway each day over 6 months.  (Our device measured sway area, and I don't have the absolute numbers any more.)

The really interesting finding was that, in our normal subject, the postural sway varied in a nice sine pattern with a period of 14 days.  We were never able to figure out what biological factor would result in a 14 day cycle.  I'd love to know.

The reason that this is relevant to your question is that the "normal" value varies over time.  If you institute a treatment at the peak of sway, and terminate at the minimum, you will think you have found a significant result, when in fact, it may just be normal variation.  Except for grossly abnormal sway, I'd be very cautious about any single-point measure.

Dear Rahul,

Previous research showed that the frequency preferably (normally) is around 0.5 Hz, but depends of the volunteers body antropometric characteristics. In general, it is observed that in the kinematic method, the inertial parameters of the body segments have considerable errors (especially in three-dimensional analysis). A simplification of this measurement method is by obtaining the COM using only a marker positioned on the fifth lumbar vertebra; but this technique can't analyze deviations in the ML direction, for example. The literature about this subject, shows that the kinematic method is the most suitable, due to its higher data reliability. In addition, the difference between the methods, if employed properly, is small. I recommend the following paper (attached).

Sincerely yours.

I agree with M.Nachiappan, however It's important to understand that COP is calculated from these ground reaction forces. In reflects the trajectory of the COM and the amount of torque applied at the support surface to control body-mass acceleration. Normally in the gait analysis, its important to measure also the COP amplitude, which to assessment potential modifications in the postural control strategy. In this way, COP velocity or speed, is the measure of the sum of displacements scalars  (ie. The cumulated distance over the sampling period) divided  by the sampling time. It represents the amount of activity required to maintain the stability.

In this way, I found some papers which deal with this subject.

I hope this helps.

Hi Liam, I guess that you will not find human studies on cervical proprioception affecting balance. I suggest you to look for studies on locomotion in rodents or rabbits. Some works of Mori et al., have suggested that certain subcortical regions in the medulla oblongata and cerebellum are responsible of different aspects in performance (possibly including some somatosensory issues). 

Hi Marcelo,

As mentioned in the other responses, there are a number of outcomes which can be used to assess the COP, as a surrogate for COM.  Most of these same outcomes can be applied directly to the movements of the COM as well.  But to your question of how to interpret the movement and speed of the COM, it depends on the lens through which you are looking.  For example, using a more traditional paradigm (i.e. linear measures such as velocity or sway) greater sway and/or greater sway velocity would be interpreted as worse balance due to the individual's inability to control their COM.  However, non-linear paradigms (e.g. time-to-boundary, entropy) view variability as a natural and needed function of a healthy organism.  Having said that, research has suggested that their may be a healthy level of variability while too little or too much may be indicative of a constrained sensorimotor system. 

I hope this helps

Erik

Marcelo,

pass me you email.

regards, A.

Postural control is the result of using the entire lower extremity (among other systems). Thus, it does not make sense to use different tests after fatigue of different LE musculature.  I would recommend using a consistent test(s) to determine the contribution that each set of muscles provides to postural stability.  For example, if postural control is negatively impacted the most after fatigue of the hip muscles, then you could conclude that those muscle contribute the most to postural stability.  However, the type of test used could also influence the results.  

The SEBT and modified SEBT have been used a lot for fatigue research but the Biodex Stability System (see link of a previous study we did) as well as single limb stance on a force plate have also been used.  Drop landing tests have also been done in the past.  

I hope this helps

Erik

Hi Jamie,

I would recommend the modified SEBT.  There is limited redundancy relative to the traditional SEBT (8 directions), and has been shown to be predictive of lower extremity injury.  While the Y and mod. SEBT measure the same construct, the mod. SEBT is more commonly used in the literature and is certainly cheaper to implement.  We exclusively use the mod. SEBT in your research studies.

I hope this helps.

Erik

My investigations of body posture have revealed that posture is highly variable and dependent upon person, activity and constraints in that activity.  More important is the capacity within the environment for anyone to change posture.  There are many techniques for measuring body posture but the ultimate question concerns the posture associated with the activity and can the posture change.  For example, short drivers use an erect posture to raise their eye height for driving while tall drivers use a slumped posture to fit in the car.  The question is whether drivers can sit in these postures in the same car and is there capacity within the car seating package to allow a change of posture.  We have learned that repetitive motions are dangerous and fixed, "good" postures are dangerous.  Thus, I would recommend a question concerning how to measure the range of postures available within an activity rather than one that assumes there is a "good posture" for everyone in the investigation.

Hi Anthony, 

please check the work of Dr. Leen Maes & Dr. Alexandra de Kegel (Ghent, Belgium), they did some interesting research on this topic.

Also, Dr Sylvette Wiener-Vacher (Paris, France) is an expert concerning dizziness in children. 

Some suggestions to read:

Ear Hear. 2014 Mar-Apr;35(2):e21-32. doi: 10.1097/AUD.0b013e3182a6ca91.

Rotatory and collic vestibular evoked myogenic potential testing in normal-hearing and hearing-impaired children.

Maes L1, De Kegel A, Van Waelvelde H, Dhooge I.

Otol Neurotol. 2014 Dec;35(10):e343-7. doi: 10.1097/MAO.0000000000000597.

Association between vestibular function and motor performance in hearing-impaired children.

Maes L1, De Kegel A, Van Waelvelde H, Dhooge I.

NeuroRehabilitation. 2013;32(3):507-18. doi: 10.3233/NRE-130873.

Evaluation and treatment of vestibular dysfunction in children.

Rine RM1, Wiener-Vacher S.

Best, 

Angelique

Lots of literature to look into in this review: 

Perhaps Tekscan has something you might find useful? 

I wrote simple program for measurement of Cobb's angle from any pictures on the PC screen. It's for Windows system. It's free on web site www.anglespine.com.

The program would be useful for you.

Excellent. You are reading in the right areas. Good luck with your study.

There are many ways to assess posture. I personally prefer the accelerometers or electomagnetic tracker located in various parts of the body  in order to assess the relationship between the body segments during the maintenance of posture and not only the sway of the body.

Dear Michael Rogers,

There are some equipments that can provide kinematic data of multiple segments of the body in order to assess postural balance besides the ones described above. You can consider the 3D electromagnetic systems (the Electromagnetic Sensors System Polhemus or the Motion Monitor electromagnetic tracking system-Innovative Sports Training, Inc., Chicago, IL, USA). Another options can be the Delsys accelerometers or the expensive and elaborated Vicon system. It will depend on your objectives. You should be aware of many aspects such as the frequency of acquisition, requirements of each system (for example, the electromagnetic systems require control of the environment regarding electromagnetic interference), and how many segments do you want to perform the kinematic evaluation. 

Hope this helps! Good luck!

For the Kinect you just have to plug it in the computer with USB. Currently it is only possible with the old Kinect (360) not the new one (Kinect One).

Then you have two options: if you use the Kinect for Xbox you have to install Kinect SDK and Windows Developper Toolkit (http://www.microsoft.com/en-us/kinectforwindowsdev/Downloads.aspx). If you use the Kinect for Windows you don't have to install anything.

For the Balance Board you simply have to connect it to the computer with Bluetooth connection.

Then it is up to you to create software to collect informations from these devices (C++, Matlab...)

The case study series was originally designed for qualitative data. His was the most advanced in terms of Hoehn and Yahr staging, of the cases I have worked with, at a 4-4.5. I used 5 standardized, well validated tools to assess symptoms of Parkinson's--motor, nonmotor, QoL. Towards the end of his life, he was unable to fill out rating scales. So I relied on my observations, and verbal reports from himself and his wife and documents his wife supplied, e.g. medication charts. My documentation is in the form of narrative SOAP notes, Does this make it mixed methods--since it became qualitative towards the end? It was not part of the design!

It is speculative if and how the interventions contributed to his balance problems and falls. He continued to do his exercises, I continued to do the massage and mind-body interventions, and he was certainly standing up straighter. He was having falls, especially in turning. My observation was that his walker was no longer set to the correct height for his postural improvement. It looked to me as though he needed to have P.T., to learn better body mechanics to adapt to his change of stature and changes of gait. My observations and opinions were not from my training as a massage therapist, but rather from my experience as a foster caregiver for elderly, and as a caregiver for my uncle who had Parkinson's. So I have a mix of perspectives and it perhaps muddied my role.

Frankly, it is still hard to be objective. I don't have a template that tells me what to do and where. Yet I recognize there may be an ethical need to report it somewhere. I wish to do so to raise the possibility that beneficial symptom changes in Parkinson's might require a medication reevaluation. If that is indeed what happened?

Previously, the Medicare guidelines required demonstrated improvement in condition to cover P.T. Some of these issues might now be moot, since the new January 2014 guidelines allow P.T. where it might improve QoL or prevent falls. A great advantage for people who have conditions such as Parkinson's.

Thank you, Susan for your sharing.

I agree with you Kesava. The sensitive nucleus of the trigeminal nerve and the limbic nuclei have many connexions with the Locus Coeruleus. The reticulospinal tract is a good idea.

Some observations I made seem to show that a certain amount of bilateral compression OR decompression of the TMJ increases the high frequencies of Heart Rate Variability at rest. These high frequencies are related to the parasympathetic power. Whereas a low decompression of the TMJ produced an increase of the high frequencies AND a decrease of the low frequencies (sympathetic/parasympathetic balance). This may suggest a complex relationship between the TMJ and the autonomic system.

Ability appropriate balance training interventions have been shown to improve postural control scores and reduce falls risk in the elderly across a number of investigations. There is also a very limited body of evidence that suggests that sensory-targeted interventions (e.g. Massage or joint mobs) have a positive effect on postural control in the elderly. There is a growing body of evidence to support this theory in individuals with ankle joint pathologies.

While not familiar with the literature, sensory reweighting studies will help connect somatosensory function with the postural control system.

All gait parameters are speed-dependent. The elderly usually select lower speeds, but this is not specific, of course. However, they tend to take shorter steps (and a higher cadence) for a given speed, compared to younger people (for various reasons). Hence, may I suggest a little and vsary simple known speed-independent parameter, combining step lenght and step cadence, i.e. the "WALK RATIO". This index is largely speed independent. The concept is that, for any given speed, the shorter the step (hence, the higher the cadence, and the lower the Walk-ratio) the more gait suffers from troubles: weakness, unbalance, pain, restricted range of joint motion, or whatever. It is really a sensitive index (although, not specific for aging but widespread in patholic conditions) and, more important, a summary of whatever troubles may affect gait. Age is a mega-trouble, of course!

See the attached paper (applying the WR to Multiple Sclerosis) for details.

My PhD is between two different schools which use the two different systems. I currently much prefer Qualisys due to the significantly reduced time to label data.

Qualisys uses an AIM model (Automatic Identification of Markers) which is basically a marker classifier which learns from each trial you track. What this means in practice is that after labelling each marker, you update the AIM model. When you open a new file, you then apply the AIM model and it will attempt to automatically label the markers over the whole trial (automatically filling in gaps of certain sizes).

Some huge benefits are when markers flicker or disappear for short periods of the movement the AIM model automatically labels it when it re-appears.

The benefit of Vicon did lie in the speed ease of which you can get kinematics and kinetics by setting up pipelines. This makes it useful for students or staff with less technical expertise. I know Qualisys have now improved their 6DOF modelling so that you can get calculate this information real time.

The school which uses Qualisys uses Visual3D for inverse dynamics. This is an extremely useful processing tool which also allows you to write automatic processing pipelines.

In terms of accuracy I am also not a fan of Vicon's "Plug in Gait" full body marker set. I require accurate knee kinetics, and therefore require an accurate knee axis. As only one marker identifies the thigh and the shank, we have found that small inaccuracies/inconsistencies with the placement of these markers can have a large impact on knee kinetics. Obviously you can use alternative marker sets with Vicon, but I thought this was still worth mentioning.