Fast WASABI post-processing: Access to rapid B and B correction in clinical routine for CEST MRI.

CEST MRI methods, such as APT and NOE imaging reveal biomarkers with significant diagnostic potential due to their ability to access molecular tissue information. Regardless of the technique used, CEST MRI data are affected by static magnetic B and radiofrequency B field inhomogeneities that degrade their contrast. For this reason, the correction of B field-induced artefacts is essential, whereas accounting for B field inhomogeneities have shown significant improvements in image readability. In a previous work, an MRI protocol called WASABI was presented, which can map simultaneously B and B field inhomogeneities, while maintaining the same sequence and readout types as used for CEST MRI. Despite the highly satisfactory quality of B and B maps computed from the WASABI data, the post-processing method is based on an exhaustive search of a four-parameter space and an additional four-parameter non-linear model fitting step. This leads to long post-processing times that are prohibitive in clinical practice. This work provides a new method for fast post-processing of WASABI data with outstanding acceleration of the parameter estimation procedure and without compromising its stability. The resulting computational acceleration makes the WASABI technique suitable for clinical use. The stability of the method is demonstrated on phantom data and clinical 3 Tesla in vivo data.