Accelerating the Whole-Brain Multi-Parametric Imaging through Joint Deep Learning Reconstruction and Physical Model Integration

Motivation: The long scan time of whole brain multi-parametric imaging limits the achievable spatial resolution and clinical application Goal(s): To develop a deep learning method that enhance the accuracy of reconstruction and quantification in 3D high-resolution multi-parametric imaging while significantly reducing scan time. Approach: This work introduced Joint DeepMTP, a multi-contrast joint deep learning model integrated with MR physical model, to accelerate the acquisition and imaging time Results: The proposed method achieved comparable reconstruction and quantification performance to the reference at 9-fold CAIPI acceleration, with a reconstruction time of 3 minutes Impact: The proposed method accelerated 3D whole-brain multi-parametric imaging while simultaneously quantifying T1/T2*/QSM/PD, benefiting clinicians with faster, high-resolution scans.