Fast T1/B1 Mapping using multiple dual TR RF-spoiled Steady-State Gradient-Echo Sequences

Introduction: Dynamic Contrast-Enhanced MRI (DCE-MRI) is widely applied to assess tissue perfusion and vascular permeability [1]. In most of these applications, a T1 map, obtained prior to contrast agent (CA) administration, is used to convert the signal intensity from a series of T1-weighted spoiled gradient-echo images into CA concentration. The accuracy of the approach is mainly affected by the flip angle dependency of the steady-state signal, which necessitates careful quantification of B1 inhomogeneities in advance [2]. This study investigates a new approach called “Multiple TR B1/T1 Mapping” (MTM), capable of fast, simultaneous B1 and T1 mapping [3]. This approach is based on the “Actual Flip angle Imaging” (AFI) sequence [4], but uses multiple TR pairs instead of the standard AFI approach of a single TR pair. Recently, MTM has been investigated in terms of B1 mapping performance [3]. In this work, MTM is analysed with respect to its T1 mapping performance in comparison with an inversion recovery reference sequence and in due consideration of the limited time allowed in a clinical set-up. Theory: The standard AFI employs a dual TR RF-spoiled steady-state gradient-echo sequence. The signal intensities S1 and S2 measured in the intervals TR1 and TR2, can be described analytically [4] (Eq. 1). The coefficient Fi is different for signals S1 and S2. For the first image S1, it is given in Eq. (2), with * 2 0 0 ~ T TE e M M − ⋅ = and 1 2 , 1 2 , 1 T TR e E − = . S2 can be obtained from Eq. (2) by interchanging indices. MTM employs multiple dual TR sequences using different TR1, TR2 in subsequent measurements. Each dual TR sequence yields two data points F1(TR1,TR2,α,T1) and F2(TR2,TR1,α,T1) from signals S1 and S2 for each pixel. The parameters α, T1, and M0 can then be obtained by a numerical fit of the theoretical expressions for the signal intensities Eqs. (1,2) to the set of measured data points. Hereby, α and T1 are obtained simultaneously and independently of each other. Thus, no further correction of either parameter has to be made, and no systematic B1/T1 errors are present in this method. Sequence parameters can be optimized to allow for optimal B1 and/or T1 mapping. In this study, the Cramer Rao Theorem (CRT) [5] was used to achieve best T1 mapping performance. Subjects and methods: (1) Sequence parameters for MTM were optimized with respect to maximum SNR in T1 mapping using CRT. For optimization, a target T1 = 900ms was chosen. SNR of MTM T1 mapping was predicted as a function of T1. (2) Experiments were conducted using a 1.5T clinical scanner (Philips Healthcare, Best, The Netherlands) on a calibrated phantom (Test Object 5, Eurospin II Test System, Diagnostic Sonar LTD). An in-plane resolution of 0.98 × 0.98 mm2 and 10mm slice thickness (4 slices) was chosen. MTM T1/B1 mapping was applied using TR11,12 = 115/750 ms, TR21,22 = 105/202 ms, and TR31,32 = 105/130 ms. An inversion recovery single shot TSE sequence (IR-TSE) served as an independent T1 measurement (TR = 10s, IR delays [ms] = 5000, 2244, 1582, 1188, 906, 686, 506, 353, 221, 105). For better comparison, acquisition times were adjusted to approximately 5 min each. (3) The same MTM sequence was tested on a healthy volunteer using the same 1.5T scanner. Results/Discussion: Predicted SNR via CRT and measured SNR for MTM T1 mapping are shown in Fig. 1. IR-TSE T1 map is shown in Fig 2a. MTM T1 and B1 maps are shown in Fig. 2b and 2c. A quantitative comparison is shown in Fig. 3. MTM T1 phantom results were found in agreement with manufacturer specifications. The maximum deviation from the true value was 126 ms for MTM and 257 ms for IR-TSE. In comparison with IRTSE, MTM results were superimposed by stronger noise. Maximum noise (1/SNR) was 11.5% for MTM and 1.4% for IR-TSE. In vivo MTM T1 and B1 results are shown in Fig. 3. Conclusion: Efficient and accurate baseline T1 and B1 quantification is a pre-requisite for standardized and clinical DCE-MRI. This work presents an approach to fast and simultaneous T1 and B1 mapping. Fitting of a theoretical model to a steady-state gradient-echo signal delivers independent T1 and B1 values. Adjusted in scan time (5min), MTM T1 mapping was found to be more accurate than IR-TSE in calibrated phantom measurements at the expense of the SNR of the delivered T1 maps. It is expected that the SNR of MTM can be enhanced by combining MTM with an EPI read out scheme. References: [1] Padhani AR, JMRI 16 (2002), 407422 [2] Treier R, et al., MRM 57 (2007), 568-576 [3] Voigt T. et al., ISMRM (2009), 4543. [4] Yarnykh VL., MRM 57 (2007), 192200. [5] Kay SM. Fundamentals of Statistical Signal Processing: Estimation Theory. Englewood Cliffs, NJ: Prentice Hall; 1993. ) , , , ( ) ~ , , ( ) ~ , , ; , , , ( 1 2 1 0 * 2 0 0 1 * 2 2 1 T TR TR F M TE T M M T T TE TR TR S i α α ⋅ = (1)