Spin and Gradient Multiple Overlapping-Echo Detachment Imaging (SAGE-MOLED): Highly Efficient T₂, T 2 * $$...

Combined spin- and gradient-echo EPI (SAGE-EPI) offers advantages in tissue quantification and dynamic imaging but suffers from low spatial resolution and geometric distortions. This study aims to develop a multiple overlapping-echo detachment-based SAGE acquisition (SAGE-MOLED) to enable efficient, distortion-corrected T₂, $T_2^{*}$ , and M₀ mapping for perfusion MRI.

SAGE-MOLED was designed as an optimized MOLED variant by refining echo time sampling and integrating multi-train blip-reversed EPI to enhance distortion correction and temporal SNR, enabling reliable extraction of subject-specific arterial input functions (AIFs). To support dynamic imaging, a steady-state Bloch simulation-based synthetic data framework was developed to simultaneously model T₁, T₂, and $T_2^{*}$ -related effects, providing training data for an end-to-end deep learning model that enables efficient multiparametric quantification. In addition, a signal-to-concentration model tailored for dynamic MOLED signals was formulated for accurate estimation of permeability and leakage-corrected perfusion parameters. The proposed technique was validated in water phantom experiments, healthy volunteers, and a pilot clinical study.

Single-shot SAGE-MOLED demonstrated high consistency with standard methods in both phantom and in vivo experiments, with Pearson correlation coefficient = 0.991 for T₂ and 0.988 for $T_2^{*}$ mapping in the brain. Compared to conventional SAGE-EPI, SAGE-MOLED mitigated geometric distortions and intravoxel dephasing-related signal loss. In perfusion MRI, dynamic SAGE-MOLED enabled simultaneous permeability and leakage-corrected perfusion parameter estimation with a single-dose contrast injection.

SAGE-MOLED overcomes key limitations of SAGE-EPI, providing high-fidelity, distortion-corrected T₂, $T_2^{*}$ , and M₀ maps for simultaneous quantification of permeability and perfusion parameters.