Neuro-oncology biomarkers allow probing tissues beyond anatomical or relaxation contrast. However, new techniques are needed to achieve mechanistic elucidation and effective therapeutic interventions. Quantitative MRI techniques, such as T1, T2, diffusion, and perfusion imaging, provide insights into tumor biology, infiltration, microstructure, and hemodynamics, and have been implemented clinically, but are limited in probing cellular and molecular changes. Saturation transfer (ST) MRI methods, including magnetization transfer contrast (MTC) and chemical exchange saturation transfer (CEST), exhibit enhanced sensitivity to semisolid macromolecular content, mobile proteins, tissue metabolites, and acidosis. Despite these advantages, ST-MRI techniques typically have been qualitative or semi-quantitative, time-consuming, and impractical for clinical implementation. However, emerging ST-MR fingerprinting (ST-MRF) represents a significant analytical advancement by integrating RF saturation preparation into a pseudo-randomized MRF acquisition framework with a deep learning‐based reconstruction. This approach enables simultaneous, clinically realistic, quantitative mapping of bulk water relaxation properties (e.g., T1 and T2), semisolid macromolecular parameters, and exchangeable solute dynamics (e.g., pool size ratios and exchange rates). We provide a comprehensive background, methodological principles, and evolving clinical applications of ST-MRF in neuro-oncology. We explore current advancements, strengths, limitations, and future possibilities for clinical translation into neuro-oncology workflows, including tumor characterization, treatment response monitoring, infiltration mapping, and biomarker development.
Quantitative MRI has been held back, because it increases scan time on top of clinical protocols. In this cross-sectional study, we explore how accurately unmeasured contrasts can be synthesized from magnetic resonance fingerprinting (MRF) data. MRF, T1-weighted (T1w), gadolinium-enhanced T1w fat-saturated (Gd T1w fs), and double-echo steady-state (DESS) data were acquired at 3 T from volunteers having knee joint symptoms (n=88). Ten subjects were excluded due to missing MRF-data. U-Nets were trained to synthesize the other contrasts out of the MRF data, and the image quality was evaluated for both the synthetic and conventional by two radiologists. Presence of synovitis was evaluated from Gd T1w fs images. Wilcoxon signed-rank test was utilized to test differences in the image quality scores. Synthetic images overall had only slightly lower image quality in radiologists’ evaluation for T1w (3.54 vs 3.88, p<0.01), DESS (3.22 vs 3.90, p<0.01), and Gd T1w fs images (3.40 vs 3.76, p<0.01). Synovitis could not be reliably observed from synthetic Gd T1w fs images. The image quality gap observed earlier persisted in this study and manifested itself steadily across all the joint tissues. Synovitis assessment was not possible from the synthetic contrast-enhanced images.
Objective While most sodium (23Na) MRI studies focus on apparent tissue sodium concentration (aTSC), longitudinal (T-1) and transverse (T-2*) relaxation times may provide additional microstructural and physiological information. This study aimed to establish reference values for 23Na T-1 and bi-exponential T-2* in human lower-leg muscle at 3 T and 7 T, evaluate methodological factors influencing T-2* quantification, and assess the reproducibility of aTSC.Methods Twenty healthy volunteers underwent 23Na MRI of the lower leg at 3 T and 7 T using 3D stack-of-stars UTE acquisitions. T-2* was quantified from multi-echo data using bi-exponential fitting with both fixed and free short-component fractions. T-1 was measured with a saturation-recovery approach. aTSC was calculated using calibration phantoms.Results T-1 increased with field strength (3 T 26.5 +/- 2.0 ms; 7 T 28.1 +/- 1.7 ms), while long T-2* decreased (3 T T(2)l* 20.5 +/- 2.9 ms; 7 T T(2)l* 18.4 +/- 2.5 ms). T-2* estimation depended on echo sampling and fitting assumptions. Mean aTSC was 17.5 +/- 1.7 mM at 3 T and 18.5 +/- 1.7 mM at 7 T, with a reproducibility coefficient of variation of 4.1%. Moderate positive correlations between field strengths were observed for short T-2*, its fraction, T-1, and aTSC.Conclusions This study provides reference values for sodium T-1 and bi-exponential T-2* relaxation in human skeletal muscle at 3 T and 7 T and demonstrates high reproducibility of aTSC across field strengths.
Neuropathies can selectively affect individual fascicles within peripheral nerves, but diagnosis is limited by MRI spatial resolution and signal-to-noise ratio. This feasibility study aimed to leverage 7 T MRI to measure sub-millimetric T2 maps of sciatic nerve fascicles in healthy subjects. Accelerated multi-echo spin–echo sequences were acquired in six healthy volunteers at 7 T. Acceleration included generalised autocalibrating partially parallel acquisition (GRAPPA) and GRAPPATINI, which integrates GRAPPA with model-based accelerated relaxometry. Extended phase graph (EPG) modelling of signal decay was used to compute T2 maps. Quantitative maps were visually assessed and T2 values were extracted from regions of interest. High-resolution imaging (0.4×0.4x3.0 mm3) enabled visualisation of individual fascicles. T2 values were lower in fascicles (57.0 ms) than in the non-fascicular intraneural compartment (76.2 ms) and intermuscular fat (132.7 ms). Compared with GRAPPA alone, GRAPPATINI reduced acquisition time (11:05 to 3:35 min) while maintaining fascicle depiction and T2 consistency (mean difference: 1.45 ms). Overlaying T2 on S0 maps enhanced fascicle delineation, facilitating anatomical interpretation. This approach demonstrated the feasibility of high-resolution T2 mapping of peripheral nerve fascicles at 7 T, offering a potential tool for improved assessment of nerve microstructure and early neuropathy detection.
