3D MR Fingerprinting for Quantitative Bladder Wall T<sub>1</sub>, T<sub>2</sub>, and M<sub>0</sub> Mapping in Healthy Subjects at 1.5 T and 3 T.
Authors
Affiliations (3)
Affiliations (3)
- Department of Radiology, University of Michigan, Ann Arbor, Michigan, USA.
- Department of Urology, University of Michigan, Ann Arbor, Michigan, USA.
- Department of Biomedical Engineering, University of Michigan, Ann Arbor, Michigan, USA.
Abstract
Current diagnostic tests for lower urinary tract symptoms (LUTS) do not assess tissue-level alterations in the bladder wall. This study assesses the feasibility of 3D magnetic resonance fingerprinting (MRF) for simultaneous T<sub>1</sub>, T<sub>2</sub>, and proton density (M<sub>0</sub>) mapping of the bladder wall in healthy subjects at 1.5 T and 3 T. A 3D MRF acquisition was combined with a deep image prior reconstruction to achieve whole-bladder T<sub>1</sub>, T<sub>2</sub>, and M<sub>0</sub> mapping in 11.8 min at an acquired voxel size of 1 × 1 × 3 mm<sup>3</sup> (interpolated to 0.5 × 0.5 × 1.5 mm<sup>3</sup>). Twenty-nine healthy subjects (14 female, 18-78 years) were enrolled, with 16 scanned at both 1.5 T and 3 T. Conventional maps were acquired using MOLLI and T<sub>2</sub>-prepared GRE sequences. Maps were analyzed by measuring mean global and regional T<sub>1</sub> and T<sub>2</sub> values, with spatial heterogeneity assessed using the coefficient of variation (CV). Bladder wall thickness (BWT) was measured from synthetic weighted images derived from MRF maps, compared to reference values from conventional T<sub>2</sub>-weighted MRI. Mean global MRF bladder wall measurements were 984 <math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mo>±</mo></mrow> </math> 53 ms (T<sub>1</sub>), 51.4 <math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mo>±</mo></mrow> </math> 5.6 ms (T<sub>2</sub>), and 4.5 <math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mo>±</mo></mrow> </math> 1.6 mm (BWT) at 1.5 T and 1468 <math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mo>±</mo></mrow> </math> 118 ms (T<sub>1</sub>), 43.5 <math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mo>±</mo></mrow> </math> 4.0 ms (T<sub>2</sub>), and 4.3 <math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mo>±</mo></mrow> </math> 1.6 mm (BWT) at 3 T. MRF T<sub>1</sub> values were significantly lower than MOLLI at 1.5 T (bias -151 ms) but not significantly different at 3 T. MRF T<sub>2</sub> values were significantly lower (by 7-8 ms) than T<sub>2</sub>-prepared GRE at both field strengths. MRF exhibited lower CV than MOLLI T<sub>1</sub> mapping at 1.5 T. BWT from synthetic images exhibited excellent agreement (bias <math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mo>≤</mo> <mn>0</mn></mrow> </math> .1 mm) with conventional T<sub>2</sub>-weighted images. Regionally, T<sub>1</sub> and T<sub>2</sub> tended to be higher in the dome and posterior wall. In an exploratory analysis, a positive association between age and T<sub>2</sub> was observed. In conclusion, 3D MRF of the bladder wall is feasible in healthy subjects, providing a baseline for translational studies to characterize pathological remodeling in LUTS.