Exercise pulley machine for MRI

[Timetraveller83]

Hi everyone, I am building a calf exercise machine for MRI. I have attached a sequence of 3 pictures showing what I have so far with the foot plate (where the patient will rest their foot) in 3 different position. As it is shown, the patient will be lying with the head to the left and their foot will rest in the left part of the machine. They will perform dorsiflexion and plantar flexion (rotating their foot forward and backward) and pushing or pulling against the footplate.

I have it setup so that the force on the footplate is translated to the disc at the right via a lever arm. The goal is to have the same rotation at the angle on the disc which is attached to a weight stack, much like at a gym machine. It's important that their be a linear translation of force from the foot to the weight stack.

Do I have the setup right? Do the radii of the footplate/ankle rotation and disc just have to be the same? You can see that the range of motion is limited with this design which is fine since patients can't rotate their feet too far in either direction, but as you reach the limits you can see that their does not seem to be a linear translation of distance rotated from the footplate to the disc. I've played with a lot of configurations with the axis of rotations for the disc and footplate at different levels and so on. Am I doing something wrong?

Thanks a lot.

 

[itsmoked]

I cannot tell being an electron herder but I thought you couldn't have any metal around an MRI. (where's my picture of a pallet jack that was sucked into the bore?)

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Timetraveller83]

You are correct, I am not using any ferromagnetic materials in the design. I will be using plastics mostly like PVC or ABS. However, metals such as brass and aluminum are not magnetized in the field so those are possibilities too, although then I begin to worry about what in MRI is called shimming. It's really a beautiful thing, MRI, but incredibly complicated.

 

[BobM3]

How are the weights being lifted? It looks like the carriage is moving on the top rail holding the weight stack.

 

[Timetraveller83]

I apologize for the misleading picture. The carriage/cart to the right is not moving, I only moved it so you could have a full view of the wheel.

I want the force from the foot transferred to through the rod to the disc/wheel to the right to then pickup the weight stack directly underneath it. So as the footplate turns the wheel, the wheel then lifts the weights stacked below.

I want to make sure that whatever distance the footplate moves or circle it inscribes is the same distance the wheel moves to pick up the weights.

However, it seems that as you reach certain position while turning the wheel, there is no long a linear relationship between how much the foot plate turns and the turning of the wheel.

Thanks for looking everyone.

 

[BobM3]

In general the force and stroke relationship is not linear. However you can get it pretty close. The forces need to be perpendicular to the moment arms to keep the relationship linear. As you rotate the pedal (and wheel) the angle between the moment arms and the forces deviate a bit from 90 degrees (sine 90 degrees not quite equal to sine 85 degrees). That's what causes the non-linear relationship.

Short strokes and long moment arms help. I'm guessing your setup as is (if the picture is scaled right) would produce less then 20% error (maybe even a lot less).

 

[BobM3]

Forget my 20% error guess. The left-most setup has the force in the link no where near 90 degrees to the moment arm so the errors would be larger.

You might be able to negate that a bit by how you attached the output weights to the wheel (a link vs. a cable around the OD of the wheel). You could also mount the link so that it is horizontal at mid-stroke and connected to the top (or bottom) of the wheel (force in link and moment arm at 90 degrees) instead of 45 degrees below the horizontal as shown in the middle picture.

 

[Timetraveller83]

Thank you that is a huge help, I didn't know how to explain what was happening before and what to aim for. I am trying out configurations according to your explanation and they are looking good, I'm going to move into manufacturing the machine now.

Thanks again

 

[SincoTC]

Timetraveller83,

You said "metals such as brass and aluminum are not magnetized in the field so those are possibilities too"

I'm not certain about brass, but significant eddy currents can be generated in aluminium near large magnetic fields so rather than being "sucked in" it could be levitated and propelled with some force! In fact this is the principle behind Maglev Trains.

 

[Timetraveller83]

I will have to check about the aluminum. I have not personally used it but have seen it published in medical literature. I test everything before it goes into the magnet and I have tested brass and it is safe. I also think lead is safe.

The metals will not be going into the magnet but used at most in the weight system. Thanks for pointing that out.

 

[MikeyP]

Aluminium doesn't get pulled anywhere, but if you start moving it in such a way that it cuts though flux lines, then huge resisting forces can be generated (on a smaller scale, this is how many home rowing machines work).

I had the opportunity a while back to play with an approx 10 cm diameter piece of aluminium plate in a relatively small (1 Tesla) MRI. Moving it up/down or left/right was easy enough. Trying to twist it was like it was submerged in molasses - and the faster you turn it, the more solid it gets.

M

--
Dr Michael F Platten