Toward Super-Resolution Reconstruction of Diffusion-Relaxation MRI Using Slice Excitation With Random Overlap (SERO)

Diffusion MRI probes tissue microstructure, but low SNR and limited resolution hinder detection of features and parameter estimates. We introduce slice excitation with random overlap (SERO), which enables variable repetition times (TRs) and diffusion weighting within a single shot. This acquisition supports super-resolution reconstruction of baseline signal ( $S_0$ ), diffusivity ( $D$ ), diffusional variance ( $V$ ), and longitudinal relaxation ( $T_1$ ) maps.

We implemented a diffusion-weighted spin-echo sequence in Pulseq that excites thick slices at random positions. Across shots, pseudo-random overlap produces inter- and intra-slice TR variation (0.15-21.9 s) with b-values up to 1.4 ms/μm². The $T_1$ -weighting enables through-slice super-resolution and allows $T_1$ estimation. Accuracy and precision were evaluated in numerical phantoms across variable SNR. SERO was compared with slice-shifting super-resolution and conventional high-resolution imaging. Feasibility was demonstrated in healthy brain in vivo at 1.5-mm isotropic resolution in 2:30 min.

In simulations SERO improved accuracy of $D$ , $V$ , and $T_1$ while maintaining voxel-wise precision comparable to direct sampling across SNRs. Regularized SERO achieved RMSE ≈ 0.5 μm²/ms ( $D$ ) and ≈ 0.5 μm⁴/ms² ( $V$ ) at SNR = 3, whereas direct sampling required SNR ≥ 7-10; root-mean-variance decreased by > 50% versus an unregularized fit. In vivo, SERO yielded sharp tissue boundaries and smooth parameter maps.

Random slice overlap enriches encoding diversity, improving accuracy and precision of diffusion and relaxation parameters without longer scan time. SERO offers a novel path to high-resolution microstructural imaging, especially at low SNR.