Rapid high‐fidelity T2* mapping using single‐shot overlapping‐echo acquisition and deep learning reconstruction

Abstract

Purpose To develop and evaluate a single‐shot quantitative MRI technique called GRE‐MOLED (gradient‐echo multiple overlapping‐echo detachment) for rapid T2*$$ {T}_2^{\ast } $$ mapping. Methods In GRE‐MOLED, multiple echoes with different TEs are generated and captured in a single shot of the k‐space through MOLED encoding and EPI readout. A deep neural network, trained by synthetic data, was employed for end‐to‐end parametric mapping from overlapping‐echo signals. GRE‐MOLED uses pure GRE acquisition with a single echo train to deliver T2*$$ {T}_2^{\ast } $$ maps less than 90 ms per slice. The self‐registered B0 information modulated in image phase was utilized for distortion‐corrected parametric mapping. The proposed method was evaluated in phantoms, healthy volunteers, and task‐based FMRI experiments. Results The quantitative results of GRE‐MOLED T2*$$ {T}_2^{\ast } $$ mapping demonstrated good agreement with those obtained from the multi‐echo GRE method (Pearson's correlation coefficient = 0.991 and 0.973 for phantom and in vivo brains, respectively). High intrasubject repeatability (coefficient of variation <1.0%) were also achieved in scan–rescan test. Enabled by deep learning reconstruction, GRE‐MOLED showed excellent robustness to geometric distortion, noise, and random subject motion. Compared to the conventional FMRI approach, GRE‐MOLED also achieved a higher temporal SNR and BOLD sensitivity in task‐based FMRI. Conclusion GRE‐MOLED is a new real‐time technique for T2*$$ {T}_2^{\ast } $$ quantification with high efficiency and quality, and it has the potential to be a better quantitative BOLD detection method.

Sub-second whole brain T2 mapping via multiband SENSE multiple overlapping-echo detachment imaging and deep learning

“…Although previous studies have considered noise as a non-ideal factor, the approach with a single direct reconstruction network is feasible only at low noise levels (That is,  x y x , e » in equation (3)). At a high acceleration factor MB, y cannot be approximated as the same as x anymore, and the single direct reconstruction network approach cannot reconstruct T 2 maps with good quality (Yang et al 2023a). Therefore, we divide the reconstruction process into two steps, and the introduction of the PnP algorithm can avoid the great impact of excessive noise on the reconstruction results.…”

Improved Resting-State Functional MRI Using Multi-Echo Echo-Planar Imaging on a Compact 3T MRI Scanner with High-Performance Gradients

“…The ME scheme has been demonstrated with robust performance in various applications, including in a line-scanning fMRI with high spatiotemporal resolutions [16], in a combined gradientand spin-echo acquisition for simultaneous quantify T 2 * and T 2 relaxation time quantification [17], and in blood flow measurements to quantify T 2 * relaxation time [18]. Recently, a deep learning model was proposed for end-to-end parametric mapping for an overlapped multi-echo sequence [19]. Although a single echo train was used, overlapping multi-echo signals can be generated with multiple excitations.…”