A strategy to reduce the sensitivity of inhomogeneous magnetization transfer (ihMT) imaging to radiofrequency transmit field variations at 3 T. Issue 3 (15th November 2021)

• Record Type: Journal Article
• Authors: Soustelle, Lucas; Troalen, Thomas; Hertanu, Andreea; Mchinda, Samira; Ranjeva, Jean‐Philippe; Guye, Maxime; Varma, Gopal; Alsop, David C.; Duhamel, Guillaume; Girard, Olivier M.
• Is Part Of: Magnetic resonance in medicine. Volume 87:Issue 3(2022)
• Page Start: 1346
• Page End: 1359
• Publication Date: 2021-11-15
• Subjects: ihMT -- inhomogeneous magnetization transfer -- magnetization transfer -- myelin -- neuroimaging -- radiofrequency field inhomogeneities; Nuclear magnetic resonance -- Periodicals; Electron paramagnetic resonance -- Periodicals; 616.07548
• DOI: 10.1002/mrm.29055
• Languages: English
• ISSNs: 0740-3194

Abstract

Abstract : Purpose: To minimize the sensitivity of inhomogeneous magnetization transfer gradient‐echo (ihMT‐GRE) imaging to radiofrequency (RF) transmit field ( B 1 + ) inhomogeneities at 3 T. Methods: The ihMT‐GRE sequence was optimized by varying the concentration of the RF saturation energy over time, obtained by increasing the saturation pulse power while extending the sequence repetition time (TR). Different protocols were tested using numerical simulations and human in vivo experiments in the brain white matter (WM) of healthy subjects at 3 T. The sensitivity of the ihMT ratio (ihMTR) to B 1 + variations was investigated by comparing measurements obtained at nominal transmitter adjustments and following a 20% global B 1 + drop. The resulting relative variations ( δ ihMTR ) were evaluated voxelwise as a function of the local B 1 + distribution. The reproducibility of the protocol providing minimal B 1 + bias was assessed in a test‐retest experiment. Results: In line with simulations, ihMT‐GRE experiments conducted at high concentration of the RF energy over time demonstrated strong reduction of the B 1 + inhomogeneity effects in the human WM. Under the optimal conditions of 350‐ms TR and 3‐µT root mean square (RMS) saturation power, 73% of all WM voxels presented δ ihMTR below 10%. Reproducibility analysis yielded a close‐to‐zero systematic bias (ΔihMTR = −0.081%) and a high correlation (ρ² = 0.977) between test and retest experiments. Conclusion: Concentrating RF saturation energy in ihMT‐GRE sequences mitigates the sensitivity of the ihMTR to B 1 + variations and allows for clinical‐ready ihMT imaging at 3 T. This feature is of particular interest for high and ultra‐high field applications.