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!

Tuesday, November 2, 2010

FOD happens!

Pieces of metal, especially magnetic ones, will find their way into all sorts of strange and potentially detrimental locations inside an MRI. During your safety training you will have learned a lot about the hazards of chairs, keys, rotary mops, oxygen cylinders and other objects that have, at one time or another, found their way into or onto an MRI – often with disastrous consequences.

Yet there is another category of foreign objects or debris - known as FOD to aviation types - that doesn’t get as much attention during safety training, largely because there are fewer safety issues. There are, however, serious implications for the quality of your data.

Finding FOD

Take yesterday, for example. There we were, a service engineer and I, rooting around in the back of the magnet checking for carbonization, testing locking nut security and the like, in a quest to identify sources of spikes that had shown up in the morning’s QA data. (I’ll do a separate post on spikes another day.) We (meaning the engineer) had already found, cleaned and replaced “standoff” spacers for the gradient power cables. These spacers – especially the one for the X coil, which gets the most use as the read axis gradient for EPI – are prone to micro-arcing, a phenomenon that can be discerned by the telltale black carbon deposits on one or both ends of the metal tube.

Anyway, we were just checking other potential spike sources when the eagle-eyed service engineer spotted a piece of stainless steel wire, no more than 8 mm in length, 3-4 mm in diameter, neatly tucked underneath the edge of a mounting bracket for the gradient tube. The FOD was extremely well hidden. After some grappling with non-magnetic needle-nosed pliers the FOD was retrieved. Now, given its location, the fact that it was snipped at both ends with wire cutters and was of a composition typical of the types of materials used in magnet and/or building construction, the chances are this particular FOD had been sitting in the bore since the MRI was installed. Was it a source of some of the spikes? Possibly. We can’t know for sure because we had other confirmed spike sources. In any event, it’s a lesson that FOD happens.

My problem?

Ah, but...! I hear you exclaim. I am just a user! I am not a physicist or an installation engineer. It is not my job to check for FOD such as this, nor am I likely to introduce pieces of metal wire under a gradient mounting bracket! Quite so, and technically you are correct. However, let’s go over two more examples that are directly pertinent to you as a regular user.

Example number two happened a couple of weeks ago. As one of my routine steps when setting up anything or anyone in the scanner I always completely disconnect and inspect both the plugs and the sockets of the RF coil I’m about to use. I look at the copper pins for discoloration – see above discussion on carbonization – or bent pins, then I gently (GENTLY!) run a finger across all the pins to make sure none are wiggly. Next, I retract the socket covers (where the coil plugs go) one at a time on the patient bed and inspect for FOD. Two weeks ago I found a tiny piece of metal wire, 5 mm long, similar to a fragment of steel wool, lurking under a socket cover. Fortunately, the little FOD hadn’t managed to work its way into the socket itself so it was easy to retrieve and discard, no harm done.

My third example happened a few years ago. A user called me in a panic: his subject had only half a head on MRI! The user assured me that the subject had had a whole head when he’d arrived, and this was verified as the subject was retrieved from the magnet. Sure enough, on the MRIs most of the occipital and parietal cortex was AWOL in some test EPIs, while on an MP-RAGE there was a pretty good ‘hole’ sans scalp signal that was rather difficult to miss.

We checked the subject’s hair for hair spray, metallic coloring, hair clips and other devices known to cause issues for MRI. We found nothing, so the subject was thanked and sent on his way. I inspected the RF head coil and was about to test it with a phantom when I ran my fingers across the rear interior surface of the coil. Ouch! I’d pricked my finger on the teeniest iron filing, one I couldn’t even see; it was embedded in the thin foam protecting the coil’s rear surface. Mystery solved.

Minimizing FOD

FOD, especially the teeny, tiny, magnetic variety, will routinely literally walk its way into an MRI suite; it gets tracked in on shoes. And unless you operate your magnet room like a hermetically sealed NASA satellite facility, FOD is gonna happen. If FOD doesn’t walk in then it will get introduced via another item of clothing. Gowning subjects only partly solves the problem: how many experimenters themselves don a clean suit in a clean room before entering the MRI facility?

So the next time you’re getting ready to scan, take a moment to inspect the magnet bore and environs. See any black dust particles gathered around the bottom of the bore? That’s FOD. Small magnetic particles such as these can be easily and quickly removed either with sticky tape or by carefully brushing the particles onto a piece of paper placed flat on the bore surface; just fold the paper up carefully with the dust inside, then bring the paper out of the magnet.

Some facilities use adhesive mats to clean shoes of debris as people enter the magnet room. It’s not a bad idea but it’s no panacea. Like many things in MRI, good procedures can usually do as well if not better than ‘technical’ solutions. For your own experiments, then, I would recommend the following standard steps at setup time:

  1. Minimize the number of shoes that walk into the magnet room. Have subjects remove their shoes before they enter the room, and check the soles of your own shoes for FOD.
  2. Check your own and your subject’s clothing for items that might become FOD.
  3. Check your own and your subject’s hair for items that might become FOD. Hair clips are a safety hazard as well as potential FOD!
  4. Check all the copper pins on the RF coil plugs for physical/mechanical flaws, then check the RF coil sockets on the patient bed for FOD.
  5. Inspect the magnet bore for all manner of FOD, from paper clips to iron dust. As a good member of your user community, remove any small FOD you find. (Large FOD? Unsure? Call your facility physicist.) Once inside the magnet there is little harm to expect from FOD that is already on the magnet bore….so long as it stays where it is. What you wish to avoid is the possibility of FOD migrating to a position where it can do more harm, such as inside an RF coil socket on the patient bed.
  6. Make a visual and tactile check of the internal surface of the RF head coil for any small particles of FOD.
  7. When the bed is run into the magnet, do a quick visual check for metal fragments the table mechanism might have spat out, or other FOD that has found its way under the bed sled.

Screening for magnetic oxygen cylinders, cell phones, wrenches and other ferrous objects continues to be of paramount importance for your own and your subjects’ safety. To ensure that you are also able to minimize the chances of magnetic stuff creating data quality issues, it’s good practice to add an extra layer of screening for the small stuff.

Remember, FOD happens!

No comments:

Post a Comment