Single‐shot T₂ mapping via multi‐echo‐train multiple overlapping‐echo detachment planar imaging and multitask deep learning

Abstract

Background: Quantitative magnetic resonance imaging provides robust biomarkers in clinics. Nevertheless, the lengthy scan time reduces imaging throughput and increases the susceptibility of imaging results to motion. In this context, a single-shot T 2 mapping method based on multiple overlapping-echo detachment (MOLED) planar imaging was presented, but the relatively small echo time range limits its accuracy, especially in tissues with large T 2 . Purpose: In this work we proposed a novel single-shot method, Multi-Echo-Train Multiple OverLapping-Echo Detachment (METMOLED) planar imaging, to accommodate a large range of T 2 quantification without additional measurements to rectify signal degeneration arisen from refocusing pulse imperfection. Methods: Multiple echo-train techniques were integrated into the MOLED sequence to capture larger TE information. Maps of T 2 , B 1 , and spin density were reconstructed synchronously from acquired METMOLED data via multitask deep learning. A typical U-Net was trained with 3000/600 synthetic data with geometric/brain patterns to learn the mapping relationship between MET-MOLED signals and quantitative maps. The refocusing pulse imperfection was settled through the inherent information of METMOLED data and auxiliary tasks. Results: Experimental results on the digital brain (structural similarity (SSIM) index = 0.975/0.991/0.988 for MOLED/METMOLED-2/METMOLED-3, hyphenated number denotes the number of echo-trains), physical phantom (the slope of linear fitting with reference T 2 map = 1.047/1.017/1.006 for MOLED/METMOLED-2/METMOLED-3), and human brain (Pearson's correlation coefficient (PCC) = 0.9581/0.9760/0.9900 for MOLED/METMOLED-2/METMOLED-3) demonstrated that the METMOLED improved the quantitative accuracy and the tissue details in contrast to the MOLED. These improvements were more pronounced in tissues with large T 2 and in application scenarios with high temporal resolution (PCC = 0.8692/0.9465/0.9743 for MOLED/METMOLED-2/METMOLED-3). Moreover, the METMOLED could rectify the signal deviations induced by the non-ideal slice profiles of refocusing pulses without additional measurements. A preliminary measurement also demonstrated that the METMOLED is highly repeatable (mean coefficient of variation (CV) = 1.65%). Conclusions: METMOLED breaks the restriction of echo-train length to TE and implements unbiased T 2 estimates in an extensive range. Furthermore, it

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

“…Compared to previous work using MOLED acquisition, 28–32 GRE‐MOLED can achieve distortion‐corrected parametric maps (Figure 5). This may be attributed to the B 0 information modulated by the multiple gradient echos with different TEs, as well as the large receptive field of R2AttUNet.…”

“…Nevertheless, because the B 0 field carried by GRE‐MOLED is also distorted, it is difficult to address signal “pileups” caused by strong local B 0 inhomogeneity (see Figure S6 for details). The development of multiple echo trains 31 with blip up–down acquisition 16 may help achieve better performance in distortion correction.…”

“…Because the acquired MR image (also called overlapping‐echo image) contains multiple $$$ {T}_2^{\ast } $$$‐weighted signals, it can be used to generate parametric maps (e.g., $$$ {T}_2^{\ast } $$$, M 0 , and ∆ B 0 ). In the absence of a 180° pulse and the limitation of echo refocusing, GRE‐MOLED allows higher temporal resolution compared to previous MOLED methods 30–35 . The scan time per slice is reduced from 170 ms (MQMOLED 35 ) to 88 ms (GRE‐MOLED), which facilitates whole‐brain quantitative BOLD FMRI.…”

“…In the absence of a 180 • pulse and the limitation of echo refocusing, GRE-MOLED allows higher temporal resolution compared to previous MOLED methods. [30][31][32][33][34][35] The scan time per slice is reduced from 170 ms (MQMOLED 35 ) to 88 ms (GRE-MOLED), which facilitates whole-brain quantitative BOLD FMRI. Note that the TE sampling range of GRE-MOLED can be flexibly adjusted to suit different scan tasks.…”

“…On the other hand, MOLED pioneers a new acceleration paradigm that differs from conventional approaches (e.g., compressed sensing or parallel imaging), that is, using echo-specific phase modulation to increase the density of available information. As a highly flexible framework, MOLED has been validated in efficient quantification of T 2 [30][31][32] and diffusion coefficient, 33 simultaneous multislice T 2 quantification, 34 and even single-shot multi-parametric quantification (MQMOLED). 35 In this study, we developed a novel qMRI technique for rapid T * 2 mapping based on the MOLED framework, termed GRE multiple overlapping-echo detachment (GRE-MOLED).…”