Sunday, April 10, 2011

CONTACT INFORMATION: Jennifer Stoffer and Novel-Pliance USA

It would be greatly remiss of me to go on without acknowledging Jennifer Stoffer and Novel-Pliance and their efforts to assist us in our research and discovery. Here is contact information for themL

Jennifer Stoffer, DVM
Gaston, OR
503-522-5529
info@backinthesaddlesite.com






novel electronics inc
Maria Pasquale, MS
964 Grand Avenue
Saint Paul, MN  55105
651-221-0505 (voice)
mariapasquale@novelusa.com
www.novelusa.com

Pressure Testing at Dr. Hilary Clayton's Lab at Michigan State University

MSU - AT LAST!

Visiting MSU was an amazing experience start to finish. It began as we drove toward the MacPhail Equine Research Facility early on the morning of April 5. The building’s façade and distinctive green roof were instantly familiar from  the group’s online presence and annual reports. I was traveling with RP’s Lisa Jordan, and before we’d even entered the building, we caught each other’s eye and grinned with delight. Some years ago the two of us took Dr. Hilary Clayton’s Equine Biomechanics class (offered through Equinology at the time) and our respect for Dr. Clayton’s intellect, commitment to the principles of research and stunning body of work was embedded in both of us during that course.

THE GOAL FOR THE DAY
We had arranged this trip in order to evaluate the Novel-Pliance sensor mat. This pad has the best reputation as a research tool – and the highest price tag – of any sensor pad on the market and we were eager to see if the result matched the build-up. We arranged to have one upper level  horse and rider, and two school horses present for the testing. All horses had their own saddles. During the course of the day we tested each horse’s saddle and then fit one or two ReactorPanel saddles to these same horses (and under the same riders). We hoped to learn as much as possible about the performance of RP relative to conventional saddles, and at the same time, to learn the capabilities of the pressure testing system so that we could see if the high cost is justified in our application. To help us in our testing, we had shipped a large box of saddles, panels, our special sorbothane rubber discs and a variety of fitting tools in advance.

THE NOW-USUAL DISCLAIMER
The following report is being made to the best of my memory, and with the understanding that I am not a scientist nor an engineer and have only a rudimentary knowledge of how the pressure sensing systems actually work. In addition, we were not permitted to video tape (my preferred method of accurately recording a testing session) and – I admit – the sequence of events were often so fascinating that I neglected to take notes.  I apologize in advance for any factual errors that I am about to commit (I’m sure there will be some). Because we were not allowed to record these sessions, I don’t have pressure sensing maps to show you; I will do my best to obtain these images and permission to put them up here in the near future.

THE NOVEL-PLIANCE PRESSURE TESTING SYSTEM
We began the morning with an overview of the system’s software capability. This session had been set up for us by the Novel-Pliance US staff who showed the utmost professionalism in response to our requests for information, and were the only one (of the four pad companies I contacted) to take our request for a demo seriously. Our experience was greatly enhanced by their presence, especially since Maria Pasquale, an engineer with a biomechanics degree and 8 years of experience at Novel, was on hand to operate the software. Unlike our last experience in which we were left wondering about some of the features of the system, in this case we were given an extremely thorough overview.
The system allows certain data fields to be entered to identify or name each record. These fields are predetermined and are not changeable. However, the next level down in the software permits quite a bit of data entry to thoroughly describe an individual testing session.

When it comes to recording data for evaluation, there are different methodologies. One is to record a certain time frame – say 15 seconds. MSU generally records a certain number of strides. In this fashion, they hope to capture the entire range of the stride in a way that permits comparison. For instance, if one was to do a time trial without counting strides, it is possible that the trial would begin just after – and end just before – the peak force moment of the stride. The longer the test, the less this would have a meaningful effect on results but since pressure tests seem to typically be conducted in short bursts, this could be a factor which impacts the end result.

The Novell-Pliance software appears to be much more sophisticated (to be fair, it might be that Maria is so well versed in operating the software that it simply seemed so).  And the superior data-crunching skills of Dr. Clayton and her assistant, LeeAnn Kaiser, added layers of meaning to the results.

DIRECT COMPARISON: THINGS OF NOTE THAT HAVE STUCK IN MY HEAD

  • The Novel pad scans at the rate of 60-80 times per second as compared to the Team-Satteltest pad at 8-15 times per second.  We don’t know how meaningful this is.
  • The Satteltest pad measures pressure; the Novel pad measures pressure and also measures force. For instance, we learned that a horse with more suspension may generate higher peak forces; this does not necessarily correlate to worse fit. In face, a saddle that permits the horse to move more freely may in fact “cause” (or enable) higher forces.
  • Both pads have 256 sensors, but the Team Satteltest pad has a significantly larger surface area. While this means the T-S pad will work under saddles with a greater weight-bearing surface, it also probably means that the distance between the sensors is greater. The Novel pad does come in a Western version which is longer in both dimensions; we did not see this pad.
  • The Novel pad is smooth; the embedded electronics are almost invisible to sight and feel. The Team-Satteltest pad has sensors which give a texture to the pad. Not sure if this is positive or not but it is a difference.
  • The Novel Pad is “sliced” front and rear so that the pad parts at the wither and again at the loin. The Satteltest pad is a single unit (though contoured for the wither). The immediate benefit to the sliced version is that shear forces of the saddle downward on the pad do not create an artificial registration of pressure on the spine (we saw this in our tests in Oregon two weeks ago). However, the ability to measure pressure on the spinal processes or spine might be very useful. Difference noted. Benefit unknown.
  • The Novel Pad can be calibrated to different scales. The generally accepted scale seems to be 60 kPA (kilo pascals - an accepted measure of force per unit area, one pascal is one newton per square meter). How the Satteltest pad is calibrated – and its scale of measurement – is so far unknown.
  • Calibrating the pad is essential for accurate results. The Novel pad has a calibration frame (at an additional cost and additional procedure). The Satteltest pad’s calibration protocol is so far unknown.
  • Unlike the Team Satteltest system that we saw last week, the Novel-Pliance system does not have a streaming mode (meaning you cannot watch the computer screen to see the pressure testing results as the rider goes around). It is possible to approximate this by recording a session without saving the result which is less convenient but perfectly functional.
  • The Team-Satteltest system includes built-in video: this is a great tool for correlating pressure to results and learning if a high pressure point occurred while the horse stumbled or the rider lost balance. The video for the Novel system is an option that adds thousands of dollars to the cost.
  • Both systems have a graphical representation of the rider’s center of pressure (this is not the center of balance, and is certainly not the center of gravity – we were mistaken earlier).


In (preliminary) conclusion - we believe that the Novel-Pliance system is fantastic in a scientific/research environment. We're not sure yet that it is the best choice for our application, which is to give clear information about saddle fit to people who do not, perhaps, understand physiology, anatomy, biomechanics or saddle construction but who are responsible for making decisions about their horses' welfare. But it is certainly an item of desire for our own use.

NEXT STEPS:
A direct comparison of sensor pads is a logical next step, so that we can learn what, if any, impact there is on accuracy when evaluating the different scan rates, sensor density, video on board, and software in a side-by-side comparison. Daring to dream, I’m imagining another testing day – perhaps back at MSU – where we gather together as many different testing systems as possible.

NEXT POST:
The horses and saddles we tested at MSU, and the results. Hopefully with pressure scans.

Sunday, April 3, 2011

The Nitty-Gritty: Our First Pressure Tests

WHAT IS THE VALUE OF A PRESSURE TEST?

Before we get into the detail of the images, it’s important to note that the scientific community does not yet agree on how much pressure under the saddle is too much. Perhaps there will never be an absolute standard due to the number of variables. For instance, what might be tolerable pressure on a horse with even muscling and pliant tissue might be excruciating on a horse with existing damage. I am hoping that we’ll find a scale of acceptability eventually . Using the computer’s results, I feel that we could possibly immediately group saddles into three categories: “quite good”, “quite bad”, and “need to correlate to other factors as the test alone is inconclusive”. This is my own opinion and it is one that I might revise after next week’s session in Michigan with the experts: Hilary Clayton and her laboratory technicians who have been using pressure testing in sophisticated studies for several years. So – please read on with the understanding that this is a process of building knowledge and understanding; we are not yet at the point that we’re able to share absolute conclusions.

PAUSE TO GIVE THE NORMAL AMERICAN SETS OF CAVEATS FOR EMPTORS:

To remind you, we conducted tests of several horses wearing conventional saddles, and then wearing ReactorPanel saddles. In each case the ReactorPanel saddle was fitted hastily, and in each case the performance of the ReactorPanel  saddle was superior based on the pressure test, and also on horse and/or rider improvement in performance. While these are extremely exciting results, this research is preliminary. The testing sample was too small to be statistically valid, and none of the horses tested were wearing professionally fit or custom conventional saddles. I hope that RP will perform as well in Michigan this week!

(FINALLY) THE PRESSURE TESTING DATA
Below, you’ll see images of the saddles we tested last week. These images are screen shots provided by Jennifer Stoffer, DVM (you’ll find her contact information at the bottom of this post). The interpretation of the day’s work is my own or based on notes I took at the time. One of the drawbacks of this technology is that some interpretation of the results is needed: the scans do not always speak for themselves.

Here they are:
Conventional saddle: bridging, skimpy weight distribution
ReactorPanel Encore with large weight bearing surface
















Above, you’re looking at two images. Each shows the left and right sides of the saddle – hopefully with a gullet channel between them. The system averages pressures among the 256 sensors and then displays different pressures in different colors. We were not told what calculations are used to provide the pressure map, nor what the gradiations in color really mean.  If we become seriously interested in this technology, these are questions we will ask the manufacturer.

You are looking at scans on a Welsh Cob who is competitive at first and second levels. The saddle on the left is the horse’s usual saddle, an older Passier that was not fit for the horse. The scan on the right shows the ReactorPanel Encore dressage saddle. Both scans are aggregates showing the averages over a trot scan of approximately 15 seconds.

This horse’s conformational highlights are:
·         Short-coupled
·         Curved topline
·         Prominent shoulders (bulging scaplulae)
·         Prominent rib cage
·         Depressed trapezius and longimussus dorsi muscles (appear damaged or even slightly atrophied)
·         Minor reaction to palpation but no areas of extreme sensitivity

The pressure test on the left revealed:

·         Weight-bearing surface is smaller than optimal
·         The saddle bridges
·         The highest pressure is at the left rear, near the spine
·         The saddle bumps into the spine at times (not apparent in this aggregate image)
·         Rider’s weight is consistently off to the left as shown by the X which marks the rider’s center of gravity
·         Front-to-back balance is 57% in front, 43% in the back

RP Encore:
·         Much broader and more even weight bearing surface
·         Does not cross spine
·         65% of weight in front
·         Highest pressure: L rear, but outside, not inside
·         Much steadier laterally (not apparent on the aggregate image)


WHEN PRESSURE TESTING PRESENTS A CONUNDRUM

The second horse presented was a 7-year-old Hannoverian mare ridden by a professional. The horse is lovely but frustrating: she resisted left flexion and would fall out of canter without warning. In this case, the pressure tests of the horse’s conventional saddle and the RP were not significantly different but in the ReactorPanel Elegance and AvantGarde dressage saddles, the mare willingly flexed to the left and sustained the canter (the owner/rider is now in the middle of a two-week evaluation of the RP AvantGarde). This points out the need to refine the testing process; in the ideal world, the computer would clearly show that one saddle was superior to the other. Here are the scans:


Conventional Saddle: heavy on the left side
ReactorPanel Elegance: Laterally balanced














In evaluating these two scans, here are the pertinent points:


  • The saddle on the left is the horse’s own Passier Grand Gilbert that was not fitted for her
  • The saddle on the right is the ReactorPanel Elegance
  • Neither saddle has a saddle pad (other than the sensor pad)
In the conventional Passier:

  • much more weight is carried to the left than the right as shown by the larger image on the left and also the rider’s “X” (showing center of gravity) that is significantly left. This is the cause of the horse’s difficulty in flexing left.
  • More of the weight is carried forward; the pressure is not even front-to-back

 In the ReactorPanel:

  • Weight is distributed over a larger area
  • Rider’s weight is more centered
  • Although the saddle appears to cross the spine, this is an artifact of the panel movement pulling on the sensor pad. In fact, the gullet channel was wide and stable.

We tested two more horses in Oregon but time prohibits posting the results at this point. I will be sure to get those images into this blog before much more time passes, but now I must sign off and pack my suitcase for tomorrow’s trip to Michigan and Dr. Clayton’s lab. I promise to take good notes and to report –as fully as I may ethically – after these sessions. As a bonus: our dear friend and advisor Dr. Kerry Ridgway is so interested in these developments that he will be meeting us in Michigan to observe our testing day. Stay tuned!