Modeling and feasibility study of table-top MRI scanner using permanent magnets

Contacts: Associate professor Kaspar Kirstein Nielsen (kaki@dtu.dk), associate professor Rasmus Bjørk (rabj@dtu.dk), senior researcher Christian Bahl (chrb@dtu.dk) or associate professor Anders Wulff (anwu@dtu.dk)

Challenge: When patients are scanned with magnetic resonance imaging (MRI), it is typically done in a full-body scanner based on superconducting magnets. These are expensive – both to purchase and to run – and in many cases only an arm, an ankle or a knee is to be scanned. In these cases, it is definitely overkill to use a full-body scanner. We therefore wish to investigate the feasibility of using permanent magnets (based on the FeNdB material) for such a table-top MRI scanner. We know that the magnetic field produced by such a system is in the range of relevance for MRI, but the size, cost and magnetic field homogeneity are so far not investigated in detail.

Idea: The idea is to use an existing model framework that is written in a combination of Matlab and Fortran. By using this system it will be possible to investigate what it takes in terms of size and cost in order to produce a specified magnetic field with a given homogeneity in a specified region of space (where the arm, knee or whatever would be placed). 

Students’ task: You will be tasked with using and perhaps further developing the already existing model. You will then design and optimize a permanent magnet system using this model and which meets the specifications of such an MRI system (in terms of a magnetic field of 1.5 tesla and a field homogeneity of no worse than 1 ppm). Once this has been developed, an economical evaluation of the system should be performed.