Magnetic resonance fingerprinting (MRF) was recently suggested for fast and quantitative chemical exchange saturation transfer (CEST) imaging. However, for in-vivo pathologies, multiple tissue parameters will vary simultaneously, thereby reducing the schedule discrimination ability and increasing the reconstruction time. Herein, we propose the sequential utilization of three MRF acquisition schedules and their incorporation in deep-learning reconstruction networks (DRONE). The technique outputs 6 quantitative maps (water, semi-solid, and amide pool properties) with acquisition and reconstruction times of 365s and <200ms, respectively. The method was evaluated in a longitudinal brain-tumor mouse study, yielding comparable parameter values to ground-truth and traditional Z-spectrum evaluations.
This abstract and the presentation materials are available to members only; a login is required.