New method for efficient eddy current characterisation
17th November 2020
Michael Schwerter, Markus Zimmermann, Jörg Felder and N. Jon Shah
In order to obtain high-quality images and data, the static magnetic field of the MRI system needs to be homogeneous. To achieve this, standard MRI systems include sets of low-order spherical harmonic (SH) shim coils that generate adjustable magnetic fields to cancel occurring inhomogeneities. However, when a patient is placed inside the MRI machine, the distribution of tissues with different magnetic susceptibilities often causes field distortions what cannot be corrected by the shim coils. Advanced methods for addressing this problem are challenging and often require shim current changes during the MR acquisition. This can cause the generation of eddy currents which further degrade the homogeneity of the magnetic field and result in images containing artefacts.
In response to this, researchers from the Forschungszentrum Jülich have proposed two innovations to existing eddy current characterization techniques, which include an efficient spatio-temporal sampling scheme and a
model-based fitting of spherical harmonic eddy current components.
It has been shown that the proposed ‘user friendly’ methods efficiently characterise and compensate for eddy current fields and can be applied to facilitate pre-emphasis implementations, such as for dynamic B0 shimming applications.
It is anticipated that, following further research and development, this innovation will be used in both clinical and fundamental research to improve the quality of data acquired in a range of MR applications.
The figure above shows the sequence diagram and sampling scheme of the proposed EC measurement approach. A) A total of NT 2D gradient-echo images are acquired during the time-course of the EC decay. The ECs induced by the shim pulse introduce phase offsets in the image data. The acquisition is repeated so as to acquire EC images in three orthogonal planes. B, An orthogonal and off-centre placement of the three imaging slices is chosen to break potential degeneracies of the EC data. The magnitude images are displayed for better visualization and illustrate the slice positioning. Abbreviations: Rx, receive; Tx, transmit
Original publication: