This study developed an SNR-efficient tensor-valued diffusion encoding (TDE) MRI protocol to enable high-resolution assessment of brain microstructure despite the longer diffusion-encoding durations required by TDE compared with conventional Stejskal–Tanner pulsed-gradient encoding. The protocol targeted 1.8-mm isotropic resolution and clinically feasible scan time by combining advanced diffusion-encoding waveforms with a multi-band multi-shot EPI readout and tailored reconstruction.
Eighteen cognitively normal volunteers were scanned at 3T: eleven older adults (5 female, 6 male; ages 62–71) and seven younger adults (3 female, 4 male; ages 22–31). Older participants underwent two scanning sessions to permit reproducibility assessment. The study received institutional review board approval (Stanford University IRB #76617) and all participants provided informed consent.
An in-house TDE sequence was implemented at 3T. To address the reduced SNR that results from the extended diffusion-encoding durations inherent to tensor-valued waveforms, the authors used an SNR-efficient multi-band multi-shot EPI readout and an appropriate reconstruction pipeline. The final protocol provided approximately 70 mm of brain coverage with 1.8-mm isotropic voxels and a total acquisition time of about 12 minutes.
The study derived multiple diffusion-derived microstructural metrics from the TDE data: mean diffusivity (MD), fractional anisotropy (FA), microscopic FA (μFA), anisotropic mean kurtosis (MKA), isotropic mean kurtosis (MKI), and total mean kurtosis (MKT). Analyses targeted three anatomical compartments: the bilateral temporal portions of the cingulum bundles (CBT), the bilateral fornix (FX), and global white matter (WM). These regions were chosen to probe white matter tracts implicated in aging and neurodegenerative processes.
Reproducibility in the older cohort was assessed using intraclass correlation coefficients (ICCs) and Bland–Altman analysis. All metrics except MKI demonstrated moderate-to-excellent reproducibility across assessed white matter regions, with reported ICCs ≥ 0.64. For group comparisons between younger and older adults, the authors used Wilcoxon signed-rank tests and applied false discovery rate (FDR) correction to control for multiple comparisons.
Exploratory comparisons between younger and older participants showed age-related differences in FA, μFA, and MKA with uncorrected p-values < 0.05, and these effects were most pronounced in the fornix. After applying FDR correction, none of the differences remained statistically significant. The authors report these findings as exploratory and interpret them cautiously given the lack of significance after multiple comparison correction.
The protocol achieved 1.8-mm isotropic spatial resolution, approximately 70 mm of axial brain coverage, and an acquisition time near 12 minutes on a 3T scanner. These parameters indicate a practical balance between spatial resolution and clinical feasibility for investigating microstructural changes in aging and potentially neurodegeneration, while partially mitigating the SNR penalty of lengthened diffusion encoding via the multi-band multi-shot EPI strategy.
The report is a preprint describing method development and an initial reproducibility and sensitivity assessment in a small cohort. Age-related differences were exploratory and did not survive FDR correction. The authors declare no competing interests. Funding sources included NIH grants (K99AG080076, R01NS095985, R01NS125781). The study followed institutional ethical guidelines (Stanford University IRB #76617). Data produced in the study are available from the authors on reasonable request.
This work demonstrates that an SNR-efficient TDE acquisition can yield sub-2-mm isotropic resolution suitable for targeted white matter microstructural assessment within a clinically feasible scan time. The reproducibility results (ICCs ≥ 0.64 for most metrics) support use of the protocol for longitudinal or group studies where high spatial resolution is required. The exploratory age comparisons suggest the protocol is sensitive to microstructural variation in small tracts such as the fornix, but larger or more targeted studies will be needed to establish robust age- or disease-related biomarkers as the observed differences did not remain significant after FDR correction.
The sequence and reconstruction were implemented in-house and used multi-band multi-shot EPI to recover SNR lost to longer diffusion encoding. The paper does not provide implementation-level parameters or reconstruction code within the preprint text; the authors state data are available upon reasonable request. Those seeking to adopt similar TDE protocols should consider scanner-specific implementation, validation of reconstruction, and careful reproducibility testing in their own environment before clinical or research deployment.