Rapid high‐resolution T1 mapping using a highly accelerated radial steady‐state free‐precession technique

• Published in: Journal of magnetic resonance imaging Vol. 49; no. 1; pp. 239 - 252
• Main Authors: Li, Zhitao, Bilgin, Ali, Johnson, Kevin, Galons, Jean‐Philippe, Vedantham, Srinivasan, Martin, Diego R., Altbach, Maria I.
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.01.2019; Wiley Subscription Services, Inc
• Subjects: Abdomen; Brain; Brain mapping; Brain slice preparation; Coefficient of variation; compressed sensing; Computer simulation; Field strength; In vivo methods and tests; Lesions; Magnetic resonance imaging; Mapping; Mathematical models; model‐based; Neuroimaging; Optimization; Personal computers; Precession; Protocol (computers); radial MRI; Reconstruction; Reproducibility; Spatial discrimination; Spatial resolution; Statistical analysis; Statistical tests; T1 mapping; Temporal resolution
• ISSN: 1053-1807; 1522-2586; 1522-2586
• DOI: 10.1002/jmri.26170

Background T1 mapping is often used in some clinical protocols. Existing techniques are limited in slice coverage, and/or spatial‐temporal resolution, or require long acquisitions. Here we present a multi‐slice inversion‐recovery (IR) radial steady‐state free precession (radSSFP) pulse sequence combined with a principal component (PC) based reconstruction that overcomes these limitations. Purpose To develop a fast technique for multi‐slice high‐resolution T1 mapping. Study Type Technical efficacy study done prospectively. Phantom/Subjects IR‐radSSFP was tested in phantoms, five healthy volunteers, and four patients with abdominal lesions. Field Strength/Sequence IR‐radSSFP was implemented at 3T. Assessment Computer simulations were performed to optimize the flip angle for T1 estimation; testing was done in phantoms using as reference an IR spin‐echo pulse sequence. T1 mapping with IR‐radSSFP was also assessed in vivo (brain and abdomen) and T1 values were compared with literature. T1 maps were also compared with a radial IR‐FLASH technique. Statistical Tests A two‐tailed t‐test was used to compare T1 values in phantoms. A repeatability study was carried out in vivo using Bland‐Altman analysis. Results Simulations and phantom experiments showed that a flip angle of 20˚ was optimal for T1 mapping. When comparing single to multi‐slice experiments in phantoms there were no significant differences between the means T1 values (P = 0.0475). In vivo results show that T1 maps with spatial resolution as high as 0.69 mm × 0.69 mm × 2.00 mm (brain) and 0.83 mm × 0.83 mm × 3.00 mm (abdomen) can be generated for 84 brain slices in 3 min and 10 abdominal slices in a breath‐hold; T1 values were comparable to those reported in literature. The coefficients of variation from the repeatability study were 1.7% for brain and 2.5–2.7% in the abdomen. Data Conclusion A multi‐slice IR‐radSSFP technique combined with a PC‐based reconstruction was demonstrated for higher resolution T1 mapping. This technique is fast, motion‐insensitive and yields repeatable T1 values comparable to those in literature. Level of Evidence: 2 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2019;49:239–252.