Synopsis

Single-voxel MR spectroscopy (MRS) serves as a reference method in fat containing tissues to probe T₁, T₂ and proton density fat fraction (PDFF). Currently, employed techniques usually rely on long TRs to minimize T₁ bias albeit the prolonged scan time and the reduced SNR efficiency. The purpose of this study is to propose the usage of a short-TR multi-inversion time (TI) multi-TE (SHORTIE) STEAM for the simultaneous assessment of T₁, T₂ and PDFF at reduced scan times when compared to long-TR multi-TI STEAM and long-TR multi-TE STEAM. The agreement between the aforementioned methods was tested in the exemplary application of adipose tissue characterization in the human supraclavicular fossa.

Introduction / Purpose

Long-TR MRS measurements minimizing $$$T_1$$$-bias are currently predominantly used in body applications investigating proton density fat fraction (PDFF) and triglyceride characteristics with applications in e.g. subcutaneous and visceral fat compartments [1] and bone marrow [2]. However, long-TR measurements have the shortcoming of prolonged scan times due to a decreased SNR efficiency. Recently, an acquisition scheme with variable TR was proposed for simultaneous $$$T_1$$$, $$$T_2$$$ and PDFF estimation in the liver.[3] Moreover, the $$$T_1$$$ sensitivity can be further improved by utilizing an inversion pulse in combination with multiple inversion times (TI) as compared to a multi-TR measurement.

To probe brown adipose tissue (BAT) multiple MR-based techniques have been proposed. [4,5] However, most of these techniques rely on the differentiation of only one criterion which – dependent on the parameter – may not always be sufficient to differentiate white adipose tissue (WAT) from BAT. [6] Therefore, the purpose of this study is to i) propose the usage of a single-voxel short-TR multi-TI multi-TE (SHORTIE) STEAM sequence enabling the simultaneous assessment of $$$T_1$$$, $$$T_2$$$ and PDFF at reduced scan times compared to long-TR multi-TI STEAM and long-TR multi-TE STEAM sequences and ii) to evaluate the measurement agreement between the aforementioned methods in the characterization of adipose tissue in the supraclavicular fossa.

Methods

Pulse sequence A single-voxel SHORTIE STEAM MRS sequence (Figure 1) was implemented with a constant minimum recovery delay $$$\tau$$$. The acquired signal as a function of the sequence parameters TI, TE, TM and $$$\tau$$$ can be described as

$$S\left( TI, TE, TM, \tau\right)=\rho\left(1-2 e^{- \frac{TI}{T_1}} + e^{- \frac{\tau+TI}{T_1}}\right) e^{-\frac{TM}{T_1}}e^{-\frac{TE}{T_2}},$$

where $$$\rho$$$ denotes the proton density signal. Quantification A joint-series time domain-based model fitting was implemented based on MATLAB's Levenberg-Marquardt algorithm. The general SHORTIE signal model can be described as

$$S\left(t\right)=e^{j \phi }\sum_i\rho_i e^{(j2\pi \omega_i - d_i )t} \left(1-2 e^{- \frac{TI}{T_{1,i}}} + e^{- \frac{\tau + TI}{T_{1,i}}} \right)e^{-\frac{TM}{T_{1,i}}}e^{-\frac{TE}{T_{2,i}}},$$

where $$$\rho_i$$$ is the proton density, $$$d_i$$$ is the damping factor also refered to as the transverse relaxation rate and $$$\omega_i$$$ is the precession frequency of the $$$i$$$th frequency component, respectively, and $$$\phi$$$ represents a common additional phase term. The fitting strategy was composed of a general plus a sequence-dependent constraint rulesets. The general constraint ruleset included the following number of variables: 5x $$$\rho$$$ (4 variables spanning a 10 peak triglyceride model [7] and a single water peak), 3x $$$d_i$$$ (1 methylene, 1 water, 1 all other fat peaks), 1x $$$\omega$$$ (common frequency shift for all frequency components), $$$\phi$$$. Sequence specific rulesets consisted of 3x $$$T_1$$$ (1 methyl, 1 all other fat peaks, 1 water) variables and 2x $$$T_2$$$ (1 fat, 1 water) variables, respectively. In vivo measurements The three MRS sequences were performed in 7 volunteers (5 male, 2 female, age: 27.2–30.8 years) in the region of the subclavicular fossa based on a water–fat imaging scout: i) multi-TE STEAM: TE=12/15/20/25ms, TR=4000ms, number of samples=4096, scan time=02:24min; ii) multi-TI STEAM: TI=10/100/500/1500ms, TE=10ms, TR=4000ms, number of samples=4096, scan time=03:00min; and iii) SHORTIE STEAM: TI=10/100/500/1500ms, TE=10/15/20/25ms, TR=742ms, number of samples=2048, scan time=01:49min. All of the three shared a common voxel position and approximate voxel-size of 9 x 9 x 9 mm³, as well as the following sequence parameters: number of averages=8, number of phase cycles=4, TM=17ms, spectral bandwidth=3000Hz. All measurements were performed on a 3T scanner (Ingenia Elition X, Philips Healthcare, The Netherlands) using the standard head coil plus anterior and built-in-table posterior coil arrays.

Results

Discussion

Conclusion

Acknowledgements

The present work was supported by Philips Healthcare.

The present work was supported by the European Research Council (grant agreement No 677661, ProFatMRI).

The present work reflects only the authors view and the EU is not responsible for any use that may be made of the information it contains.

References

[1] Gavin Hamilton et al. “In vivo triglyceride composition of abdominal adipose tissue measured by 1H MRS at 3T”. In: Journal of Magnetic Resonance Imaging 45.5 (May 2017), pp. 1455–1463. [2] Dimitrios C Karampinos et al. “ Quantitative MRI and spectroscopy of bone marrow”. In: Journal of Magnetic Resonance Imaging 47.2 (Feb. 2018), pp. 332–353. [3] Gavin Hamilton et al. “ In vivo breath-hold 1H MRS simultaneous estimation of liver proton density fat fraction, and T1and T2 of water and fat, with a multi-TR, multi-TE sequence”. In: Journal of Magnetic Resonance Imaging 42.6 (Dec. 2015), pp. 1538–1543. [4] Houchun Harry Hu. “ Magnetic Resonance of Brown Adipose Tissue: A Review of Current Techniques.” In: Critical reviews in biomedical engineering 43.2-3 (2015), pp. 161–181. [5] Dimitrios C Karampinos et al. “Techniques and Applications of Magnetic Resonance Imaging for Studying Brown Adipose Tissue Morphometry and Function”. In: Handbook of Experimental Pharmacology. Berlin, Heidelberg: Springer, Berlin, Heidelberg, 2018, pp. 1–26. [6] Gavin Hamilton et al. “MR properties of brown and white adipose tissues”. In: Journal of Magnetic Resonance Imaging 34.2 (Aug. 2011), pp. 468–473. [7] Gavin Hamilton et al. “In vivo characterization of the liver fat 1H MR spectrum.” In: NMR in Biomedicine 24.7 (Aug. 2011), pp. 784–790.