External magnetic field inside MRI scanner during scan

[killerpiraat]

Hello, i am currently doing a project for school.

In the hospital there is a MRI scanner (7T). During the MRI scan, there will be a couple of ADC's placed inside the bore, next to the patient. Those ADC's are powered externally with twisted pair wires with normal power supplies. Now it is my projectsgroup task to make the powersupply of those ADC's wireless. We will attach coils to the outside of the bore and attacht the ADC on the inside of the bore at the same location. We will use inductive coupling.

The question is: What kind of interference will there be if we are not carefull?  What is the effect of a external magnetic field INSIDE the core of an MRI scanner during a scan?

[T3sl4co1l]

Do semiconductors usually work under that field strength..?

What do you mean "bore"?  The internal cylinder, around which are wound the various gradient coils?  Where exactly will you find a free spot on that?  (Is there normally unused area?)

It should be possible with air-cored resonant coupling, but be weary of induced fields from the gradient coils, and from the superconducting magnet if anything should move within it (eddy currents).

And probably a zillion other things related to medical products, or research and ethics if this is a student/research project, etc.  Or uh, even just the warranty on that stupid thing, I can't imagine the manufacturer will like anyone gluing things onto their precise and very expensive machine?

Tim

[killerpiraat]

With bore i mean the tunnel were the patient is in during the scan.
You are absolutely right. The coil for wireless power will be placed on the outside of the internal cylinder. The placement of our coil will need a place on the core, but the contracter told us that that problem is not within the scope of this project. Our task is to make a inductive coupling systems so that the powersupply to the ADC's is transmitted through the bore. But you have a good point.

Your third line: "It should be possible with air-cored resonant coupling, but be weary of induced fields from the gradient coils, and from the superconducting magnet if anything should move within it (eddy currents)."

What will happen with the induced field? I assume you mean that the gradient coils will induce a field inside our coil? And what do you think that the effect on an MRI scan will be, if we create an magnetic field inside the MRI scanner during a scan? Im not sure if i understand that last part correctly. If what moves within the superconducting magnet? You mean if anything moves inside the bore?

PS: The MRI scanner we will be testing on is designed for research purpose. No patient are scanned in it. Offcourse we have to be carefull, we will look out for that . We will discuss it with them before we glue them on haha

EDIT: As for the semiconductors, i assume that you mean that we will need those for the rectifing part  on the receiving part of the wireless power system?

[JohnG]

The Larmor frequency for a 7 T machine will be on the order of 300 MHz for hydrogen, assuming that's the main nuclear species you are looking at (since it's a research machine, this may not be true). This matters because any harmonics in the range of this frequency, or in the range of one of the possible IF frequencies, will show up as an artifact in the image. MRIs are extremely sensitive to this, and scan rooms are typically Faraday cages to keep out outside interference. Also, the peak envelope power of the RF pulses used can be 10s of kW. You do not want to have any coils pick up anything near one of these frequencies during transmit pulses.

For example, take a look at this: https://patents.google.com/patent/US8829905B2/en

And, 7T is high. If you move your head too fast putting stuff in the bore, you may get dizzy from the eddy currents induced in your head. I assume you have had high field magnet safety training.

Cheers,
John

Hope this helps,
John

[TimFox]

When I worked in MRI years ago, our big problem with electronics near the region of interest for imaging was ferromagnetic material in the components, especially lead frames for the ICs and discrete semiconductors.  As a test, we taped an SOT-23 transistor to the side of a test object (10 cm cube of saline solution in plastic) and found a big black spot in the image.  The industry was no help:  when I asked a capacitor manufacturer about ferromagnetic materials, he assured me:  "Don't worry, our leads are pure nickel."  Extreme uniformity of the main magnetic field (7 T for your example) is critical for imaging.

[RobBarter]

Not sure if it's been mentioned above but be very careful when installing the device.  If its in anyway ferrous it will become a very fast projectile straight through the centre of the bore.
Also, if you personally have had any metal surgically fitted to you (broken leg etc) and it's not titanium then you shouldn't go into the room.   If you've been a welder in a past life....don't go into the room (Welders can have metal splinters in their eyes).

[JohnG]

This is very true. I suggest getting a big rare-earth magnet and see if it can pick up or move any of your components. If it moves them even a tiny bit, they will disrupt the MRI image. You may also be very surprised at just how many parts have some magnetic material, usually the nickel barrier plating on component terminations. Check everything that goes in the bore..

John

[JohnG]

If you've been a welder in a past life....don't go into the room (Welders can have metal splinters in their eyes).

Not just welders, but machinists as well, or really any sort of metal worker. This is usually a question you will get asked before getting an MRI, and if the answer is yes, sometimes they will do a CT first to look for something.

Note that any metal can disrupt the image due to eddy currents generated due to the gradient magnetic fields, the latter being dynamic. This is also a big problem for metal splinters, even non-magnetic ones, because they can be subject to a lot of mechanical force due the the gradient fields, and this can cause them to move about. Depending on where in your body they are, this can be a serious problem.

John

[T3sl4co1l]

Even tattoos can create image distortion, and discomfort(!).  Specifically, magnetite based blacks I guess?

So you can see, there are a lot of good reasons why experiments are usually done as remotely as possible, e.g. using plastic tube to carry audio or other pneumatic or hydraulic signals, or fiber optics (don't forget you can get multi-fiber image transmitting cable, too) to carry light or other electromagnetic signals.

Tim

[JohnG]

A very strange and interesting demo I saw was when someone held a wrench (bronze, I think, definitely non-magnetic) in a 3T MRI bore and dropped it onto the exam table. It took several seconds to fall about 30 cm through the air due to eddy current braking. Really eerie to watch.

John

[RobBarter]

A very strange and interesting demo I saw was when someone held a wrench (bronze, I think, definitely non-magnetic) in a 3T MRI bore and dropped it onto the exam table. It took several seconds to fall about 30 cm through the air due to eddy current braking. Really eerie to watch.

We did the same with a 30cm long block of aluminium letting it topple over on the couch in a 1.5T MRI.  Very weird.
This thread is very timely, reminding me of the safety rules as I'm in the shelter Friday for the first time since March.  Have some software to test.

[TimFox]

Back in my MRI days, we installed an experimental amplifier in a 2.0 T magnet.  We had carefully avoided ferromagnetic components (we used ceramic capacitors without nickel barrier, that required silver-bearing solder to avoid leaching the Pd-Ag metal), mounting the board inside an aluminum-extrusion enclosure with brass screws, a reasonable distance from the imaging volume.  I was worried when first mounting the box inside the bore as I felt a strong force on the box, but quickly realized that it was an eddy-current reaction due to the motion of the box, with no steady-state attraction or repulsion.  (As opposed to a tweezers that I absent-mindedly set down on the patient couch, which went flying into the bore.)

[JohnG]

Even brass screws are suspect. I knew someone who spent days chasing down an artifact in an MRI, and finally traced it to some brass bolts. It turns out that he could pick them up with a regular strength magnet. Apparently, brass contains a lot of recycled metal of all sorts these days, and a bit of iron is not uncommon. So, if you are expecting brass, you should check and make sure it's not "brass".

John

[TimFox]

We kept a speaker magnet next to the assembly bench to check parts--didn't have much problem with brass screws, but we bought from a good vendor.  Never trust "non-magnetic" stainless steel:  it work hardens to become quite magnetic.

[JohnG]

There has been a big push in MRI to try and reduce cost, increase reliability, and increase the number of receiver channels. The cabling has become a major problem for both cost and reliability. At least in the US, this is all regulated to the nth degree, and the signals are small, so high performance cables are needed. The combination of these factors results in high cost. Hence, the push to move more electronics from the equipment room into the scan room (inside the Faraday cage), and move the receivers, including the high speed A/Ds into the bore, with a minimum number of cables out.

There is also a push to move power electronics into the scan room, or even on the magnet. Even though there are places on the outside of the magnet where ferrite won't saturate, it's not allowed because of safety concerns around maintenance. Interference is another big problem, but one way to address that is the patent I referenced above.

John

[dmills]

The fibre as an optical power link is discussed in the AofE X chapters, there may be a vendor name in there.

I pretty much cannot think of a nastier environment for inductive power coupling, and yea, nickel barrier layers will have you unless you are VERY picky about your components, even ENIG board plating is right out!