4283

Opportunities for improved myocardial first-pass perfusion imaging at 0.55T

Ye Tian¹, Sophia X. Cui², Parveen Garg³, and Krishna S. Nayak¹
¹Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, United States, ²Siemens Medical Solutions, Los Angeles, CA, United States, ³Division of Cardiovascular Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA, United States

Synopsis

Keywords: Myocardium, Perfusion

Myocardial first-pass perfusion (FPP) MRI is used in the clinic to diagnose coronary artery disease but has limited spatial resolution and coverage. Contemporary low-field scanners have the potential to provide FPP imaging with improved resolution and coverage by leveraging different acquisitions, such as spiral bSSFP. In this study, we demonstrate a time-efficient spiral bSSFP FPP pulse sequence at 0.55T and compare it with the standard Cartesian FPP in 4 healthy volunteers. Measured SNR, CNR, and myocardial boundary sharpness scores are all higher in the spiral images than in Cartesian images.

Introduction

Myocardial first-pass perfusion (FPP) MRI is a useful tool for the diagnose of coronary artery disease. There has also been increasing interest in applying cardiac MRI on contemporary 0.55T scanners^1-4. Such systems have reduced off-resonance, different intrinsic relaxation properties (shorter T1 and longer apparent T2), and relaxed SAR constraint, which give greater latitude in pulse sequence design and the possibility of improved image quality. Non-Cartesian sampling provides higher acquisition efficiency and can simultaneously provide finer spatial resolution and broader coverage^5-7, and is favorable at low field².

In this study, we implemented a saturation recovery spiral bSSFP pulse sequence for myocardial FPP, that provides high-resolution myocardial perfusion images without any acceleration, and compared it with conventional Cartesian FPP in healthy volunteers at rest.

Methods

Pulse Sequence: Figure 1 illustrates the spial bSSFP pulse sequence. The sequence was implemented using the RTHawk (HeartVista, Inc., Menlo Park, CA) scanning platform⁸. The sequence images three slices per heartbeat, with each slice comprised of a non-selective saturation recovery preparation⁹ and 2D spiral bSSFP readout. The Cartesian sequence uses the same saturation recovery pulse followed by 2D Cartesian bSSFP with parallel imaging acceleration. Detailed scan parameters are listed in Table 1.

Protocol: Experiments were performed using a whole body 0.55T system (prototype MAGNETOM Aera, Siemens Healthineers, Erlangen, Germany) equipped with high-performance shielded gradients (45 mT/m amplitude, 200 T/m/s slew rate). Four healthy volunteers (age 21-25, 2 female) were enrolled in this study, under a protocol approved by our Institutional Review Board, after providing written informed consent. The imaging protocol includes a cardiac localizer, pre-contrast T1 mapping, rest FPP, whole-heart cine, rest FPP, post-contrast T1 mapping and LGE. In each FPP scan, 0.05mmol/kg (half dose) of gadobenate dimeglumine (MultiHance, Bracco Diagnostics, Milan, Italy) was manually injected through IV at a rate of 3-5ml/s following by a 20ml saline flush. ECG-gated FPP sequence acquired data for 60-100 heartbeats in three short-axis slices covering the basal, mid, and apical LV myocardium. In each subject, two injections were performed with a ~10-minute gap in between. In two subjects, we ran the Cartesian FPP first, and in the other two subjects we ran the spiral FPP first. The total scan time per subject was ≤ 1 hour.

Evaluation: SNR units reconstruction¹⁰ was performed offline with CG-SENSE for spiral and GRAPPA for Cartesian. SNR, CNR, and myocardial wall sharpness score as described by Lim et al.¹¹ were calculated. SNR was calculated and reported for three time points: 1) baseline LV blood and myocardium, 2) peak LV blood enhancement, and 3) peak LV myocardium enhancement. The myocardium was segmented according to the 17-segment model¹² and SNR was calculated in each segment and averaged across 16 out of 17 segments (3 short-axis slices). CNR was calculated between the peak and baseline LV blood and myocardium, respectively. Boundary sharpness scores were calculated from each cardiac segment, including endocardial and epicardial boundaries. To obtain signal enhancement curves, images were registered to a low-rank approximation with similar image contrast¹³and using LV blood and myocardium contours.

Results

Figure 2 shows perfusion images from one representative subject where spiral FPP was performed first. Both spiral and Cartesian images capture the contrast dynamics clearly; we can see the contrast enhancement in the RV blood, LV blood, and myocardium, and all myocardial boundaries are clearly depicted. We observe a small dark rim artifact in the middle slice of the Cartesian images but not on the spiral images. The spiral images appear sharper, largely due to the reduced voxel size. Figure 3 shows the signal enhancement curves from the mid-ventricular spiral image of another representative subject. We observe the signal curves in the myocardium enhance uniformly, as is expected for healthy volunteers.

Table 2 reports the quantitative SNR, CNR, and boundary sharpness scores. Spiral and Cartesian images have comparable baseline SNR in both the LV blood and myocardium. The peak myocardial and blood SNR, as well as the CNR are higher in the spial images. The boundary sharpness scores are also higher in the spiral images.

Discussion

Compared with the conventional Cartesian FPP images, the experimental spiral FPP images generally had finer spatial resolution, higher SNR, CNR, and boundary sharpness scores. We were surprised that Cartesian FPP had lower SNR, despite having significantly larger voxel size. This could be caused by the need for parallel imaging for cartesian imaging, and/or the fact that the spiral acquisition samples more densely at the k-space origin. Spiral FPP had a higher boundary sharpness, which is not surprising, since the voxel dimensions were significantly smaller.

Future studies are needed to compare 0.55T spiral FPP sequence against a conventional field strength (1.5T or 3T) reference, and to determine the feasibility of perfusion quantification. Future studies also need to evaluate the spiral FPP in patients with known CAD. Combining the spiral sequence with simultaneous multi-slice or 3D acquisitions^6,7,14,15 is also of interest as a means to provide greater spatial coverage.

Conclusion

We have demonstrated the feasibility of performing myocardial first-pass perfusion MRI at 0.55T with a time efficient spiral bSSFP. This approach simultaneously provides superior spatial resolution, SNR, CNR, and boundary sharpness compared to conventional Cartesian counterparts.

Acknowledgements

We acknowledge grant support from AHA Grant #903839, NIH R21-HL159533, and NSF Grant #1828736. We also acknowledge research support from Siemens Healthineer. We thank Mary Yung for research coordination and Elizabeth Arevalo for nursing support.

References

1. Bandettini WP, Shanbhag SM, Mancini C, McGuirt DR, Kellman P, Xue H, et al. A comparison of cine CMR imaging at 0.55 T and 1.5 T. Journal of Cardiovascular Magnetic Resonance. 2020;22(1):37.

2. Restivo MC, Ramasawmy R, Bandettini WP, Herzka DA, Campbell-Washburn AE. Efficient spiral in-out and EPI balanced steady-state free precession cine imaging using a high-performance 0.55T MRI. Magnetic Resonance in Medicine. 2020;84(5):2364-75.

3. Campbell-Washburn AE, Ramasawmy R, Restivo MC, Bhattacharya I, Basar B, Herzka DA, et al. Opportunities in Interventional and Diagnostic Imaging by Using High-Performance Low-Field-Strength MRI. Radiology. 2019;293(2):384-93.

4. Tian Y, Cui SX, Lim Y, Lee NG, Zhao Z, Nayak KS. Contrast-optimal simultaneous multi-slice bSSFP cine cardiac imaging at 0.55 T. Magn Reson Med. 2022.

5. Wang J, Yang Y, Weller DS, Zhou R, Van Houten M, Sun C, et al. High spatial resolution spiral first-pass myocardial perfusion imaging with whole-heart coverage at 3 T. Magnetic Resonance in Medicine. 2021;86(2):648-62.

6. Tian Y, Mendes J, Wilson B, Ross A, Ranjan R, DiBella E, et al. Whole-heart, ungated, free-breathing, cardiac-phase-resolved myocardial perfusion MRI by using Continuous Radial Interleaved simultaneous Multi-slice acquisitions at sPoiled steady-state (CRIMP). Magn Reson Med. 2020.

7. Yang Y, Meyer CH, Epstein FH, Kramer CM, Salerno M. Whole-heart spiral simultaneous multi-slice first-pass myocardial perfusion imaging. Magn Reson Med. 2019;81(2):852-62.

8. Santos JM, Wright GA, Pauly JM. Flexible real-time magnetic resonance imaging framework. Conf Proc IEEE Eng Med Biol Soc. 2004;2004:1048-51.

9. Kim D, Gonen O, Oesingmann N, Axel L. Comparison of the effectiveness of saturation pulses in the heart at 3T. Magn Reson Med. 2008;59(1):209-15.

10. Kellman P, McVeigh ER. Image reconstruction in SNR units: A general method for SNR measurement†. Magnetic Resonance in Medicine. 2005;54(6):1439-47.

11. Lim Y, Lingala SG, Narayanan SS, Nayak KS. Dynamic off-resonance correction for spiral real-time MRI of speech. Magn Reson Med. 2019;81(1):234-46.

12. Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, et al. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart. A statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Int J Cardiovasc Imaging. 2002;18(1):539-42.

13. Xue H, Brown LAE, Nielles-Vallespin S, Plein S, Kellman P. Automatic in-line quantitative myocardial perfusion mapping: Processing algorithm and implementation. Magn Reson Med. 2020;83(2):712-30.

14. Tian Y, Mendes J, Pedgaonkar A, Ibrahim M, Jensen L, Schroeder JD, et al. Feasibility of multiple-view myocardial perfusion MRI using radial simultaneous multi-slice acquisitions. Plos One. 2019;14(2).

15. Mendes JK, Adluru G, Likhite D, Fair MJ, Gatehouse PD, Tian Y, et al. Quantitative 3D myocardial perfusion with an efficient arterial input function. Magn Reson Med. 2019.

Figures

Figure 1. ECG-gated saturation recovery spiral bSSFP pulse sequence. The acquisition is ECG triggered. Three slices (basal, mid, and apical short axis) are imaged per R-R interval for heart rates up to 110 beats per minute. For each slice, a saturation pulse is played, followed by spiral bSSFP acquisition. A 5 TR linear ramp-up catalyzation is used to reduce transient oscillations.

Table 1. Spiral and Cartesian FPP scan parameters. This table summarizes the scan parameters used for the Cartesian and spiral FPP in volunteer 1 and 4. In volunteer 2 and 3, the major change in scan parameters was the effective Cartesian accelerations, which was 1.7 and 3, respectively. A smaller acceleration provides better image quality at a cost of longer imaging window, and a higher acceleration suffers from low SNR. With the more time-efficient spiral sampling all images were fully sampled. The voxel size in the Cartesian sequence for volunteer 3 was also decreased to 1.97 x 1.56 mm².

Figure 2. (Animated GIF) Cartesian and Spiral FPP images at 0.55T from one representative subject (volunteer 4). (a) Cartesian FPP and (b) Spiral FPP. In this subject, spiral FPP was performed first. The spiral images are sharper by appearance. Dark rim artifact is visible in the mid short-axis slice inferoseptal segment of the Cartesian acquisition, but not appear in any other location. Both sequences capture the contrast enhancements in the right ventricle, left ventricle, and the myocardium. The video is set to be played at 17 frames/second.

Figure 3. Representative time-signal intensity curves from spiral FPP of the mid-short-axis slice in one subject. Signal enhancement curves are plotted in the LV blood and 6 myocardium segments. The baseline signal intensity (averaged across the first 10 frames) were subtracted from the curves. The signal curves in the myocardium enhance uniformly as expected.

Table 2. Quantitative SNR, SNR, and boundary sharpness scores. We report the average scores for Cartesian and spiral images. We also report the scores for each subject since there were major difference in Cartesian scan parameters (acceleration and voxel size, see Table 1) and the injection order (stated as 1st or 2nd) was different. Baseline SNR was comparable between spiral and Cartesian. Spiral images have higher peak myocardial and blood SNR and CNR despite the voxel size is smaller. The boundary sharpness scores are higher in spiral images, presumably due to the smaller voxel size.

Proc. Intl. Soc. Mag. Reson. Med. 31 (2023)

4283

DOI: https://doi.org/10.58530/2023/4283