Shielding Design

Magnetic resonance imaging (MRI) has been one of the most exciting innovations in medicine in the last 50 years. The ability to visualize, at high resolution, soft tissue inside the human body has revolutionized many forms of treatment. The magnet inside an MRI device needs to produce a spherical volume of very homogeneous magnetic field, with a diameter of up to 0.5 metres or so for whole body scanners. The higher the magnetic field, the better the spatial resolution and potentially improved signal-to-noise ratio and contrast-to-noise ratio, and these fields can only be achieved using superconducting coils. With its pedigree in accelerator magnets Opera was soon grasped by the MRI magnet community and has been used in the design of virtually all MRI magnets in regular use today.


However, this also introduces a serious issue. Even the best designed superconducting MRI magnets with active shielding coils will also produce substantial stray field. In public areas, legislation mandates that DC fields should be less than 5 Gauss (0.5 mT), which usually requires the MRI to be enclosed within a shielded room made from steel sheets. Higher field magnets, which give better resolution images, only exacerbate this. Ten years ago 1.5 T magnets were the norm, now it is 3 T.

Designers of shielded rooms, consequently, have a multi-objective problem to solve. They need to ensure that fields outside the shield meet legislative requirements while seeking to minimize the amount of material used (directly for cost reasons, but also the weight of the shield may be prohibitive). In addition, the magnet manufacturer also has to assess that the shield has not introduced perturbations to the homogeneity of the image volume field that are too large to be “shimmed out” on site. Rather than relying on “cut and try” and risking over-engineering, many companies have turned to the Opera software – renowned for achieving the PPM accuracy required for MRI design – to design the shielded room.


A more difficult challenge comes when the shielded room designer does not know the configuration of the superconducting coils. This is increasingly common as new generations of MRI are introduced (3 T replaces 1.5 T), but the older models are sold on for re-use in another hospital. The shielding designer is presented with a map of the stray field from the unshielded magnet and, to successfully use Opera for shielding design, they first need to determine a set of coils which produce a similar stray field pattern. This is what is known as an Inverse Problem – what is the source of the results or measurements that we have?


In our webinar on shielding design we show how:

  • the Opera Optimizer may be used for shielding design, even when the magnet is an unknown.
  • the “thin plate” technique may be used to increase efficiency without compromising accuracy.
  • the Biot-Savart field calculation available in post-processing aids in shield design.

Opera FEA Engineering Manager Chris Riley talks about the problem in a special blog post.

In order to help out some of our customers who may not be experts in the field of Finite Element Analysis, our Application Engineers have created an automated script to guide the user through the process outlined above. Please feel free to contact us if you would like a demonstration of the automated process in action.