Ultrafast acceleration of Synthetic MRI acquisition while maintaining quantitative values: Validation study

Synthetic MRI (SyMRI) is widely used in both clinical and research settings; however, its relatively long acquisition time of approximately 5-6 min has limited its applicability to acute neurological conditions in which fast imaging is critical. The aim of this study was to validate an ultrafast SyMRI acceleration algorithm and to identify quantitative parameters that maintain high agreement with conventional protocol even under such ultrafast acquisition conditions.

Ten healthy volunteers (6 females and 4 males, median age: 27) were included. The multi-dynamic multi-echo (MDME) sequence at 3.0 T MRI were performed. Three different scan durations were used, and two types of acceleration factors-SENSE and Smart Speed AI (SSAI)-were applied, resulting in a total of three imaging protocols: 5.5-min SENSE, 4-min SSAI, and 1-min SSAI protocols. ROI measurements were performed at 15 points for each image, and quantitative values including R1, R2, T1, T2, and PD were compared. The standard deviation (SD) within the measured ROIs was also compared. Bland-Altman analysis, linear mixed-effect model, and intraclass correlation coefficient (ICC) analysis were performed.

Overall, no significant differences were observed between the acquisition methods for R2, T2, and PD. When evaluated separately across the 15 anatomical regions, no significant differences were observed between the acquisition methods for R2 and T2. The SD within the ROIs decreased with shorter acquisition times. Inter-reader agreement for the measurements was high.

R2 and T2 values showed high agreement even with acceleration in the ultrafast SyMRI with a 1-min acquisition time.