Education, tips and tricks to help you conduct better fMRI experiments.
Sure, you can try to fix it during data processing, but you're usually better off fixing the acquisition!

Saturday, November 6, 2010

FOD for thought.

As my scanner is down at the moment, with service engineers tearing things apart to identify some sources of spikes, there’s a bit of a delay in getting the resting state data I promised a couple of weeks ago. Please standby, normal service will be resumed shortly. For the time being I thought I’d continue the topic of ‘foreign objects and debris.’ Changing tacks a little bit, away from the insidious, small stuff, I thought it might be edifying to take a look at the big stuff – the stuff with major safety implications.

If you’re a regular fMRIer then you will already have been treated to safety videos demonstrating the sorts of bad things that can happen to a watermelon or a brick wall when magnetic objects are allowed to impact an MRI magnet. If you’re an fMRI newbie, welcome. May I suggest you spend a few minutes checking out YouTube videos for your enlightenment? Here are links to some goodies:

Oxygen cylinder 1 – 0 Watermelon

The rear wall of an MRI suite gets a good bashing from an oxygen cylinder

One wonders whether someone had been sitting on this chair when it started to move…

More watermelon abuse

And here is a video of some tests we did with an old 4 T magnet that was about to be decommissioned. We did a chair, too, but ours was deliberate:

Fun, eh? Sure, this stuff is exhilarating when it’s intentional and controlled. But I bet you don’t fancy being the person responsible for stabbing your subject repeatedly with the pair of scissors that you accidentally carried into the magnet room.


  1. Nice video Ben, wish I had of thought to have this sort of fun when we decommissioned our 4T in NM. Do you mind if I use these videos in some safety Training I will be doing here in Bangor shortly?


  2. G'day, Paul! Assume you are under a few feet of snow in N Wales at the mo...? Just like Cairns, eh?

    Of course you may use the video. It's also on YouTube (search for "4 T magnet safety").

    Interesting note that isn't 100% obvious from this video: the gradient peaked just inside the bore liner, where the wrench attaches itself. Past that the gradient decreased a lot, to the point where it was quite easy to move the remains of the stapler inside the bore, up to about two feet from the outer edge. At that point the forces were huge. (We tried getting the stapler out using a plastic chair as a backstop, but the stapler just punched through it! We ended up using a leather work glove to hold it, then two of us were able to pull it out.)

    Anyway, I mention the gradient because I think it will be similar for an actively-shielded 3 T. When I do the "rolled up dollar bill" demo it accelerates down the patient bed to just inside the bore, then stops. I've not tried something more magnetic than a paperclip to know whether there's a "bubble" of max. force just inside the bore, but I'd wager good money there is. And of course this has major implications for removing magnetic items, as previously noted.

    Cheers mate! See you in Quebec.

  3. One additional comment about the two "bubbles" of max. field gradient, one at either end of the magnet bore (for those who aren't familiar with such things). The existence of these two "bubbles" is what causes a magnetic object to shoot into the magnet, almost exit the back (though some objects may actually exit!), and then oscillate back and forth until the object finally attaches to the bore. The particular oscillatory characteristics are determined by the mass, mass of ferromagnetic components, cross-sectional area and the gradient profile along the bore. Thus, instead of just shooting in and sticking to the center, which is what you might expect (if there were a central peak of magnetic field gradient rather than two peaks, one at either end), a magnetic object becomes a flying, stabbing, crushing missile in the bore for quite a time.

  4. Cheers Ben. The bubbles illustrate another of the potential unforeseen problems that Active shielding can introduce, and well worth mentioning!