Inversion‐Recovery‐Prepared Oscillating Gradient Sequence Improves Diffusion‐Time Dependency Measurements in the Human Brain
Haotian Li,Tao Zu,Yi‐Cheng Hsu,Zhiyong Zhao,Ruibin Liu,Tianshu Zheng,Qing Li,Yi Sun,Daihong Liu,Jiuquan Zhang,Yi Zhang,Dan Wu
DOI: https://doi.org/10.1002/jmri.28311
IF: 4.4
2022-06-22
Journal of Magnetic Resonance Imaging
Abstract:Background Oscillating gradient diffusion MRI (dMRI) enables measurements at a short diffusion‐time (td), but it is challenging for clinical systems. Particularly, the low b‐value and low resolution may give rise to cerebrospinal fluid (CSF) contamination. Purpose To assess the effect of CSF partial volume on td–dMRI measurements and efficacy of inversion‐recovery (IR) prepared oscillating and pulsed gradient dMRI sequence to improve td–dMRI measurements in the human brain. Study Type Prospective. Subjects Ten normal volunteers and six glioma patients. Field Strength/Sequence A 3 T; three‐dimensional (3D) IR‐prepared oscillating gradient‐prepared gradient spin‐echo (GRASE) and two‐dimensional (2D) IR‐prepared oscillating gradient echo‐planar imaging (EPI) sequences. Assessment We assessed the td‐dependent patterns of apparent diffusion coefficient (ADC) in several gray and white matter structures, including the hippocampal subfields (head, body, and tail), cortical gray matter, thalamus, and posterior white matter in normal volunteers. Pulsed gradient (0 Hz) and oscillating gradients at frequencies of 20 Hz, 40 Hz, and 60 Hz dMRI were acquired with GRASE and EPI sequences with or without the IR module. We also tested the td‐dependency patterns in glioma patients using the EPI sequence with or without the IR module. Statistical Tests The differences in ADC across the different tds were compared by one‐way ANOVA followed by post hoc pairwise t‐tests with Bonferroni correction. Results In the healthy subjects, brain regions that were possibly contaminated by CSF signals, such as the hippocampus (head, body, and tail) and cortical gray matter, td‐dependent ADC changes were only significant with the IR‐prepared 2D and 3D sequences but not with the non‐IR sequences. In brain glioblastomas patients, significantly higher td‐dependence was observed in the tumor region with the IR module than that without IR (slope = 0.0196 μm2/msec2 vs. 0.0034 μm2/msec2). Conclusion The IR‐prepared sequence effectively suppressed the CSF partial volume effect and significantly improved the td‐dependent measurements in the human brain. Evidence Level 1 Technical Efficacy Stage 1
radiology, nuclear medicine & medical imaging