T1 corrected multipeak T2*-IDEAL gradient-echo imaging for the quantification of intermuscular adipose tissue

Introduction: Intermuscular adipose tissue (IMAT) has been recently associated with different metabolic abnormalities including obesity and insulin resistance [1]. The distribution and the total amount of IMAT have been traditionally assessed using T1-weighted imaging [1]. However, quantification of IMAT based on T1-weighted imaging is limited by partial volume effects and requires effective segmentation procedures [2]. IMAT fat fraction maps can be alternatively measured using chemical shift-based water-fat separation techniques [3]. IDEAL has been recently implemented to measure fat content in dystrophic skeletal muscle [4]. In order to get accurate fat content measurements with IDEAL, the spectral complexity of fat, T2* decay and T1 effects have to be considered. Multi-peak T2* IDEAL based on 6 time-point measurements has been proposed to overcome the two first confounding factors [5]. Regarding the T1 effect, the large T1 difference (values at 3 T) between muscle (1500 ms) and fat (360 ms) can cause a significant overestimation of fat fraction [6, 7]. The use of small flip angle (SFA) has been proposed to minimize T1 weighting [6, 7] and multiple flip angle (MFA) measurements have been used to correct for T1 weighting [8, 9]. However, the SFA method is limited by low SNR and the MFA method suffers from prolonged scan times [6, 9]. In the present work, the use of a precalibrated T1-corrected fat spectrum is proposed in order to correct for T1 effects in dual flip angle (DFA) multi-peak T2* IDEAL measurements of IMAT, considering accuracy and noise performance. Materials and Methods: Model formulation: The multi-peak T2* IDEAL signal model [5] is enhanced by using a precalibrated multi-peak fat spectrum that takes into account the T1 weighting of the individual fat peaks. If ap,cor are the T1-corrected relative amplitudes of the fat peaks, the enhanced signal model can be written as: