Z-spectral fitting for fat-correction of APT/CEST-MRI including fat-suppression pulses to improve body applications

Fat signal correction remains a challenge in APT/CEST-MRI for body-oncology. We show that signal background models that include direct water saturation, MT-effect, and modified fat spectra can be fitted to Z-spectra acquired with/without fat-suppression. APT/CEST signals are then extracted as the residual from the background fit (APT#). The model for fat-suppressed spectra uses a mirrored fat contribution. After B0-correction, 5 model parameters are sufficient. This allows for the acquisition of clinically feasible protocols using less than 20 Z-spectral offsets. Fat-suppression potentially increases the precision of APT# imaging by lowering the initial fat contribution.