A 3D two-point method for whole-brain water content and relaxation time mapping: Comparison with gold standard methods

Quantitative imaging of the human brain is of great interest in clinical research as it enables the identification of a range of MR biomarkers useful in diagnosis, treatment and prognosis of a wide spectrum of diseases. Here, a 3D two-point method for water content and relaxation time mapping is presented and compared to established gold standard methods. The method determines free water content, H2O, and the longitudinal relaxation time, T1, quantitatively from a two-point fit to the signal equation including corrections of the transmit and receive fields. In addition, the effective transverse relaxation time, T2*, is obtained from an exponential fit to the multi-echo signal train and its influence on H2O values is corrected. The phantom results obtained with the proposed method show good agreement for H2O and T1 values with known and spectroscopically measured values, respectively. The method is compared in vivo to already established gold standard quantitative methods. For H2O and T2* mapping, the 3D two-point results were compared to a measurement conducted with a multiple-echo GRE with long TR and T1 is compared to results from a Look-Locker method, TAPIR. In vivo results show good overall agreement between the methods, but some systematic deviations are present. Besides an expected dependence of T2* on voxel size, T1 values are systematically larger in the 3D approach than those obtained with the gold standard method. This behaviour might be due to imperfect spoiling, influencing each method differently. Results for H2O differ due to differences in the saturation of cerebrospinal fluid and partial volume effects. In addition, ground truth values of in vivo studies are unknown, even when comparing to in vivo gold standard methods. A detailed region-of-interest analysis for H2O and T1 matches well published literature values.