Accurate and precise determination of T1 values is of central importance in clinical studies and for tissue segmentation based on the myeloarchitecture that transcends T1 [1]. However, image segmentation based on MP2RAGE [2,3] may be hampered by a non-uniform transmit field, especially at 9.4T [4]. Correction for read-out and inversion efficiency, based on T2 dependent variation, enables generation of synthetic images with improved properties for image segmentation [4]. Here we investigate whether well-described age-dependent changes [5] can be detected by high field T1 relaxometry, and how different correction methods influence the results.(21-56y N=25, 6 females), who volunteered to participate in the ERB approved, were scanned at 9.4T (Siemens Germany), with a 16ch transmit/31ch receive array [6]. The transmit field was mapped with the Actual Flip Angle method [7] and T1 with the MP2RAGE sequence (TI1/TI2=900/3500ms; FA=4/6°; read-out TR =6ms; inversion TR=8894ms, 0.8mm isotropic voxel size). T1 calculations was performed pixel wise in three ways: A: without any corrections; B: correction for the actual read-out flip angle only; C: correction for transmit field dependent variation of read-out and inversion efficiency assuming a T2 of 38ms. Synthetic MP2RAGE contrast images were used for tissue segmentation in SPM12, and DARTEL based adaptation of the AAL brain atlas [8] to native space. Uncorrected and corrected T1 values, as well as flip angle scaling factors, as indicators of the uniformity of the transmit field, were extracted for each individual in the 116 brain regions inside voxels with a grey matter probability ≥0.5 (ROI). Linear regression analysis of the age-dependent variation in T1 was performed and ROIs with p<0.001 (Bonferroni corrected for multiple comparisons) were considered significant.The actual flip angle averaged within ROIs varied between 20%-165% of the prescribed angle (Fig 1, top row) so the intrinsic bias correction of the MP2RAGE method failed in brain regions with the highest transmit field deviations, leading to altered T1 values in corresponding areas (middle row, blue line). The actual transmit field correction of the read-out (middle row, green line) lead to decreased T1 values, that were further normalized by correction for variations in the inversion efficiency (red line). The coefficient-of-variation for T1 across subjects decreased across the methods and allowed a more reliable detection of age-related changes in T1. For the linear regression approach, an increasing number of ROIs for which significance was reached, N, was observed across the methods, with N=11, 47 and 61 for methods A, B, and C respectively. In the latter 61 ROIs, the annual decrease in T1 time was -2.6ms±1.7 (A); -3.6ms±0.8 (B); and -4.6ms±0.7 (C), respectively. Assuming that the age-related changes depend solely on iron accumulation [5], we found an average in vivo, r1 relaxivity of iron of 4.2±0.8 s-1/mg iron/ g wet weight tissue in cortical areas.The intrinsic bias correction of the MP2RAGE technique is not sufficient to achieve reliable quantification of T1 at ultra high magnetic fields. But, provided that a correction for transmit field inhomogeneity is performed, T1 maps that consistently reveal age-related changes can be generated. The technique holds promise for investigation of local myeloarchitectonics for neuroscientific and clinical studies.