Building an MRI safe video camera (e.g. webcam)

Question

I would like to use a small camera inside an MRI. The eye-tracking cameras that are commercially available are both too expensive and also not suitable for my purposes.

How would one go about making a small camera that is safe for an MRI?

For example, could a shielded box be built to house a regular webcam?

Would the camera need to be built from specific materials?

Additional requirements for the camera (without providing too many specifics of my actual project): -the camera should be able to record an up-close view of a patient's skin (~5cm x 5cm)

-normally the area of skin recorded is upper thigh, but can be other locations

-the camera should respond to illumination (eg fiber optic end glow lighting or similar) to help highlight skin textures

-Needs to be video recording at least 20fps

-Changes to location of interest are often very slow and difficult to observe in real time, so video recordings are reviewed later

-Should be decent resolution, at least 720p but higher is better

Open to any ideas to make this work.

Answer 1

TL;DR: there is a reason why often endoscope or fibre optic cameras are used - I recommend to stick with those solutions.

Long story: There is no way to tell without knowing the exact application, e.g. MRI and sequences specs. A co-PhD worked on MRI artefacts and safety for active implantable devices and main outcome is: there are few scenarios were the size and placement of the MRI artifact is acceptable. Not to mention the effect of stronger magnetic fields on the electronic itself.

The mentioned "shielding" mainly counters electrical and varying magnetic fields. It typically cannot prevent the penetration of the static magnetic field. One would require (perfect) diamagnetic materials (e.g. high mu metals or superconductor). In any case, any kind of shielding or magnetically susceptible material leads to sever distortion of the RF signals used for the imaging (e.g. see https://doi.org/10.1148/rg.2018170135 figure 3 b and c). Even if you find a place to position your camera such that the artifact does not interfer with your area of interested, the safety and functionality of the camera as active device is questionable. There is a high interest in having cameras inside of MRI and considering the amount of workforce/PhDs/research projects contributed to this matter is a strong indication that there is no "easy solution" - besides the already mentioned "divide and conquer" aka fibre optics etc.

EDIT: this is of course just the information I picked up during discussions with my former colleague.

Answer 2

This answer is a bit off-topic for the site, but here goes.

Note that in comments you specify that the commercial product that you have isn't suitable for your use, but I'll point out that the company you point to has a fairly large catalog in this space. Chances are actually fairly good that going that route will, in the long run, save you resources in some combination of time and money.

If you insist on going your own way, you have some homework to do. At some point in time, before there were custom solutions, people were doing this on their own. You need to do a search of your own literature, and maybe some neighboring literatures, to find out what people were successfully using. Every published paper will tell you, probably down to a part number, the products and sources that enabled the research in the paper, in the methods section. You may need to go back to a point in time before commercial products were available.

After you've done this homework, and found the parts you'd like to use, you have more homework to do, which is working with your MRI safety people to establish that what you propose to use is safe in the MRI. If you don't use a commercial product made for MRI use, you must not skip this step -- plenty of people have been hurt or killed by stuff that didn't meet the right standards flying around in an MRI suite. I can't recall which standard you need to meet, but rest assured your MRI safety people know it well. IIRC, it involves, at least in part, suspending the object from a string, applying a very strong magnet in the perpendicular direction, and measuring the angle that results.

There is probably no way that building a camera from the ground up is the approach you should be seeking, if you expect it to work in a reasonable amount of time.

Answer 3

Keep all conductors out of the bore of an MRI magnet.

Otherwise, you will distort the magnetic fields used during the scan, ruining the imaging process. There is more to the scanner than the (typically 1.5 T) static magnetic field. Much smaller gradient magnetic fields are superimposed upon the static field to select various planar sections in the scanning volume. The gradient fields are changed electronically, allowing the entire scanning volume to be interrogated and a series of images to be constructed.

When you place a conductor inside the scanner, you create a distortion of the magnetic field. It does not matter whether the material is ferromagnetic, paramagnetic, diamagnetic, etc.; any conductor can distort the magnetic field to a magnitude much greater than the gradient fields. The result is a distorted image.

This is an important reason why you are asked if you have any metal inside your body before you are given an exam. Yes, safety is always the first concern. However, even materials that are "safe" inside the scanner can still disqualify you from an MRI, because it interferes with the images. (Safe materials that are outside the scan volume are often allowed.) If you can't produce a proper image, why is your patient in an MR scanner in the first place?

So don't place your camera inside the bore. Leave it outside and optically couple it to the area you wish to record.

(Source: I used to design CT and MR scanners for General Electric.)