Instrumented Treadmills

[Shawn Farrokhi]

Hi Folks,
We are in process of evaluating a couple of different instrumented treadmills for our new Biomechanics Laboratory. We are currently comparing the dual-belt Bertec treadmill to the Tandem desing treadmill by AMTI. I’m intrigued by the Tandem design treadmill because of its capability to also act as stationary platforms. I would appreciate any insight or expertise from those who have experience using such equipment. Thank you in advance for your time.

Regards,

Shawn Farrokhi, PT, PhD, DPT
Assistant Professor
Department of Physical Therapy
Department of Bioengineering

[Daniel Leib]

Re: Instrumented Treadmills

Hi Shawn,

I don't have experience with the side-to-side treadmills, but I do have experience with AMTI's tandem design. Overall the treadmill works quite well, but there are of course some issues to consider. The data coming out of the treadmill will be very noisy compared to in-ground plates. From the data I've seen it's not a big deal for walking, but for running the forces in the medio-lateral directions can be pretty nasty due to the belt slipping over the plates.

Aside from the noise, though, usability is pretty smooth. Their supplied control software works quite well and the plates are easily integrated into a Vicon (or other) motion capture system as they're immaculate about documenting both electrical and distance measurements. When I was working at Ball State we worked with Forrest from AMTI to get it set up and he was very helpful. Delivery of the treadmill was several months late, however, due to on-the-fly revisions in the design process. This was about two years ago though so this may not be as much of an issue if they've worked out some of those kinks.

Hope this helps,

Dan

[Manvendra Singh]

Re: Instrumented Treadmills

Also check http://www.forcelink.nl/index.php?op...d=13&Itemid=20 . They specialize in Instrumented Treadmills, and have robust, accurate treadmills designed for both walking and running analysis including very large, custom built ones for special applications.

[Oshri Even-zohar]

Re: Instrumented Treadmills

And V-Gait from MOTEK MEDICAL here

[Shawn Farrokhi]

Re: Instrumented Treadmills

Thanks Dan. I have also heard from several other people that the medio-lateral forces could be very noisy.

Do you have any experience using the force platforms with the belt not running (i.e. for static activities)? I assume the forces would be much cleaner during activities such as squatting or just balance activities.

[Jason R. Franz]

Re: Instrumented Treadmills

Hello Shawn,

We successfully used a tandem design AMTI instrumented treadmill at the University of Virginia to collect treadmill and overground force data (treadmill acting as static force plates) in:

Riley et al., A kinematic and kinetic comparison of overground and treadmill running. MSSE 2008; 40(6):1093-100.

The forces we obtained from the instrumented treadmill without the belts running were just as clean as those obtained from typical static force plates. You'll notice in the article though that more filtering was required for the treadmill forces to obtain similar signal to noise characteristics as those collected overground.

Best,

Jason Franz
Doctoral Student
Department of Integrative Physiology
University of Colorado Boulder

[Eric Walker]

Re: Instrumented Treadmills

Hi Shawn,

Currently, we are using a split-belt instrumented treadmill system from Bertec. I have also had some brief experience with the ADAL3D system from medical development. Based on initial set up and operation, I prefer the Bertec system over the ADAL3D. Both companies were very helpful with set up, and the technology on the ADAL system may have improved since my experience with them about 5 years ago.

The belt movement does introduce more noise into the recordings compared to static forceplates, but it can be easily filtered out. In our Bertec system, the majority of the noise appears to be mainly from belt movement, and not due to electrical noise from the motor. I do not know if this is unique to Bertec, since there is some initial processing of the signals within the force plate.

Good Luck,
Eric

[Matthew Seeley]

Re: Instrumented Treadmills

Hi All,

I write to ask the same question that was asked by Shawn one year ago. Except, I am wondering if anyone has anything new to add to the existing replies (which have been helpful). I am wondering if there have been any significant changes in the world of instrumented treadmills in the past year. Thanks.

Matt

[Ross Miller]

Re: Instrumented Treadmills

Hi Matt,

We got the "Compact" model of the AMTI tandem treadmill recently. It had a strange issue with a "knocking" sound that synched with the drum rotations and moved around all over the place (sometimes in the front, sometimes back, left, right, changed in volume) but AMTI came out immediately and worked on it on-site and replaced a few parts, and it seems to be working fine now.

The signals as expected are quite noisy. For the vertical channel I have to filter the force data at ~ 30-50 Hz to get a curve approaching the "smoothness" of an overground vertical GRF (for one trial anyway; it's probably preferable to filter at a higher rate and collect lots of trials, and hope that averaging smoothes things out). For walking I think cutoffs this low are probably okay, but they would probably present some issues for studying running, especially if you're interested in impacts.

We haven't done any comparisons between the overground and treadmill data so I can't comment on that. My perceptions from reading papers on the topic (e.g. the Riley et al. study cited earlier) is that the faster you go, the more dis-similar the overground and treadmill data become.

I've never used a split-belt model but I've been on them before and always felt like I was doing a bit of a "cowboy" gait (bow-legged), particularly when running.

Ross

[Scott Dueball]

Re: Instrumented Treadmills

I'm surprised that no one has identified what is the most obvious difference: cost. Sure there are limitations to both designs but more often than not the significant difference in cost is the deciding factor. If you are seriously considering these than I expect you already know that. Both companies (Bertec and AMTI) do a great job promptly servicing the equipment. The noise is a factor in both devices. Depending on the capture software you are using, raising the minimum force requirement is the quickest method but may not be the best depending on the application. There are articles describing intricate filtering methods if you are looking for that.

I installed MoCap systems with force plates and played around with both treadmills. That said, I would always prefer to buy an AMTI if I had the funds. A number of labs I worked with had bought treads with grand plans to use them but never really got the research going. The only people I have worked with who regularly used the treads had the AMTI.

[Robert D. Catena]

Re: Instrumented Treadmills

We have a Bertec split belt. Regrading Ross's comment about the bow-legged gait, we have found that with this design:
- walking for elderly is fine
- walking for healthy young individuals often results in bad trials where the subject hits the same belt with different feet. We just perfer to use our in-ground plates for this.
- Running with shoes on is fine
- Running with no shoes will require you to move the person over one belt because people are reluctant to step on the split portion without shoes on. I don't blame them either.

I've never used a split-belt model but I've been on them before and always felt like I was doing a bit of a "cowboy" gait (bow-legged), particularly when running.

Ross