Highly Accelerated SSFP Imaging with Controlled Aliasing in Parallel Imaging and integrated-SSFP (CAIPI-iSSFP) - PubMed (original) (raw)

Highly Accelerated SSFP Imaging with Controlled Aliasing in Parallel Imaging and integrated-SSFP (CAIPI-iSSFP)

Thomas Martin et al. Investig Magn Reson Imaging. 2017 Dec.

Abstract

Purpose: To develop a novel combination of controlled aliasing in parallel imaging results in higher acceleration (CAIPIRINHA) with integrated SSFP (CAIPI-iSSFP) for accelerated SSFP imaging without banding artifacts at 3T.

Materials and methods: CAIPI-iSSFP was developed by adding a dephasing gradient to the balanced SSFP (bSSFP) pulse sequence with a gradient area that results in 2π dephasing across a single pixel. Extended phase graph (EPG) simulations were performed to show the signal behaviors of iSSFP, bSSFP, and RF-spoiled gradient echo (SPGR) sequences. In vivo experiments were performed for brain and abdominal imaging at 3T with simultaneous multi-slice (SMS) acceleration factors of 2, 3 and 4 with CAIPI-iSSFP and CAIPI-bSSFP. The image quality was evaluated by measuring the relative contrast-to-noise ratio (CNR) and by qualitatively assessing banding artifact removal in the brain.

Results: Banding artifacts were removed using CAIPI-iSSFP compared to CAIPI-bSSFP up to an SMS factor of 4 and 3 on brain and liver imaging, respectively. The relative CNRs between gray and white matter were on average 18% lower in CAIPI-iSSFP compared to that of CAIPI-bSSFP.

Conclusion: This study demonstrated that CAIPI-iSSFP provides up to a factor of four acceleration, while minimizing the banding artifacts with up to a 20% decrease in the relative CNR.

Keywords: CAIPI-bSSFP; CAIPI-iSSFP; CAIPIRINHA; Extended phase graphs (EPG); Simultaneous multi-slice (SMS).

PubMed Disclaimer

Figures

Fig. 1

Fig. 1

Pulse sequence diagrams for CAIPI-bSSFP (left) and CAIPI-iSSFP (right) and CAIPI-iSSFP with flow compensation gradients. CAIPI-bSSFP and CAIPI-iSSFP were the same except for the added dephasing gradient in the readout direction, where it caused a 2Π dephasing of the spins within a voxel, which resulted in an averaging of the bSSFP signal profile. To compensate for and reduce flow related artifacts and signal loss in CAIPI-iSSFP, bipolar gradients were added. CAIPI-bSSFP = controlled aliasing in parallel imaging and balanced steady-state free precession; CAIPI-iSSFP = controlled aliasing in parallel imaging and integrated SSFP; RF = radiofrequency; SMS = simultaneous multi-slice; TR = repetition time

Fig. 2

Fig. 2

Comparison of bSSFP, iSSFP, and SPGR images without CAIPIRINHA acceleration methods. The iSSFP contrast was more similar to that of bSSFP than SPGR. White arrows indicate a blood signal that was visible in bSSFP and iSSFP, but not SPGR. Blue arrows highlight the banding artifacts in bSSFP image that were not present in iSSFP and SPGR. bSSFP = balanced steady-state free precession; CAIPIRINHA = controlled aliasing in parallel imaging results in higher acceleration; iSSFP = integrated SSFP; SPGR = spoiled gradient echo

Fig. 3

Fig. 3

Numerical simulations plotting the signal profiles of bSSFP, iSSFP, and SPGR as a function of the flip angle (top) and off resonance (bottom) for white matter (WM), gray matter (GM), liver, and venous blood (VB). The simulated signal for iSSFP was less than bSSFP and higher than SPGR for all tissue and blood, and was not sensitive to off-resonance. bSSFP = balanced steady-state free precession; iSSFP = integrated SSFP; SPGR = spoiled gradient echo; TR = repetition time

Fig. 4

Fig. 4

Comparison of CAIPI-iSSFP and CAIPI-bSSFP with SMS acceleration factors of 2 (a), 3 (b), and 4 (c). Examples of unaliased images from a single simultaneously excited slice are shown. The phase modulation for each of the images is shown above. There were banding artifacts clearly present in b) and c) in the phase modulations of 0 and Π/2 respectively, while in the CAIPI-iSSFP images the banding artifacts were not visible. The image contrast between the different tissues was comparable between the two sequences, and the noise did not significantly degrade the image quality with an SMS factor of 4. CAIPI-bSSFP = controlled aliasing in parallel imaging and balanced steady-state free precession; CAIPI-iSSFP = controlled aliasing in parallel imaging and integrated SSFP; SMS = simultaneous multi-slice

Fig. 5

Fig. 5

Example images showing SMS factors of 2, 3, and 4 with CAIPI-iSSFP (bottom) and CAIPI-bSSFP (top) sequences using a long TR of 8.4 ms. There were severe banding artifacts in the CAIPI-bSSFP images; whereas, the CAIPI-iSSFP images did not have banding artifacts. CAIPI-bSSFP = controlled aliasing in parallel imaging and balanced steady-state free precession; CAIPI-iSSFP = controlled aliasing in parallel imaging and integrated SSFP; SMS = simultaneous multi-slice

Fig. 6

Fig. 6

Plots showing the measured relative CNR of white matter and ventricles (a), and white and gray matter (b) for bSSFP, iSSFP, and SPGR with CAIPIRINHA acceleration SMS = 2, 3, and 4. The relative CNR was less for iSSFP compared to bSSFP, but had a much higher relative CNR than SPGR. As the SMS increased, the relative CNR decreased due to the g-factor. bSSFP = balanced steady-state free precession; CAIPIRINHA = controlled aliasing in parallel imaging results in higher acceleration; CNR = contrast-to-noise ratio; iSSFP = integrated SSFP; SMS = simultaneous multi-slice; SPGR = spoiled gradient echo

Fig. 7

Fig. 7

CAIPI-bSSFP and CAIPI-iSSFP with SMS factor of 3 and iSSFP with flow compensation liver scan. The blue arrows identify the banding that was present in the CAIPI-bSSFP images. Spinal fluid and VB signal was suppressed in CAIPI-iSSFP due to the spoiling. However, the banding artifacts were removed and the aorta still showed a bright signal. Adding the M1 flow compensation improved the spinal fluid and VB signal (red arrows). CAIPI-bSSFP = controlled aliasing in parallel imaging and balanced steady-state free precession; CAIPI-iSSFP = controlled aliasing in parallel imaging and integrated SSFP; SMS = simultaneous multi-slice; VB = venous blood

Fig. 8

Fig. 8

Comparison of bSSFP, iSSFP, iSSFP with M1 flow compensation, and SPGR sequence images without CAIPIRINHA in the liver. The VB signal in the iSSFP and SPGR images was significantly less due to the spoiling. However, the banding artifacts were removed compared to bSSFP. The flow compensation did improve the VB ghosting artifacts; however, some of the VB signal within the liver was not as bright. bSSFP = balanced steady-state free precession; CAIPIRINHA = controlled aliasing in parallel imaging results in higher acceleration; iSSFP = integrated SSFP; VB = venous blood

References

    1. Miller KL, Tijssen RH, Stikov N, Okell TW. Steady-state MRI: methods for neuroimaging. Imaging Med. 2011;3:93–105.
    1. Carr H. Steady-state free precession in nuclear magnetic resonance. Phys Rev. 1958;112:1693–1701.
    1. Bieri O, Scheffler K. Fundamentals of balanced steady state free precession MRI. J Magn Reson Imaging. 2013;38:2–11. -PubMed
    1. Scheffler K, Lehnhardt S. Principles and applications of balanced SSFP techniques. Eur Radiol. 2003;13:2409–2418. -PubMed
    1. Oppelt A, Graumann R, Barfuß H, Fischer H, Hartl W, Schajor W. FISP: a new fast MRI sequence. Electromedica. 1986;54:15–18.

LinkOut - more resources