Email updates

Keep up to date with the latest news and content from Journal of Cardiovascular MR and BioMed Central.

This article is part of the supplement: Abstracts of the 2011 SCMR/Euro CMR Joint Scientific Sessions

Open Access Open Badges Poster presentation

Highly-Accelerated Real-Time Cine MRI using compressed sensing and parallel imaging

Li Feng, Ricardo Otazo, Monvadi B Srichai, Ruth P Lim, Daniel K Sodickson and Daniel Kim*

  • * Corresponding author: Daniel Kim

Author Affiliations

New York University School of Medicine, New York, NY, USA

For all author emails, please log on.

Journal of Cardiovascular Magnetic Resonance 2011, 13(Suppl 1):P25  doi:10.1186/1532-429X-13-S1-P25

The electronic version of this article is the complete one and can be found online at:

Published:2 February 2011

© 2011 Feng et al; licensee BioMed Central Ltd.

This is an open access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


To develop and evaluate highly-accelerated real-time cine MRI using compressed sensing and parallel imaging.


Breath-hold cine MRI with balanced steady-steady free precession (b-SSFP) may yield non diagnostic image quality in patients with impaired breath-hold capacity and/or arrhythmias. In such patients, it may be necessary to perform real-time cine MRI. Currently, dynamic parallel imaging methods, such as TSENSE [1] and TGRAPPA [2], can be used to achieve only moderate acceleration rates (R) of 2-3 using standard body and spine coil arrays. We propose the application of a recently developed joint acceleration technique (CS-PI)[3] that combines compressed sensing [4] and parallel imaging for highly-accelerated, real-time cine MRI with clinically acceptable spatiotemporal resolution.


Real-time cine MRI pulse sequences with b-SSFP readouts and TGRAPPA and CS-PI accelerations with R=4 and R=8 were implemented on 3T whole-body MRI scanners (Siemens; Tim-Trio & Verio) equipped with standard body and spine coil arrays (12 elements total). The relevant imaging parameters include: FOV=320mm x 320mm, acquisition matrix size=128x128, TE/TR=1.37/2.7ms, receiver bandwidth=1184 Hz/pixel, and flip angle=40o. The temporal resolutions were 86.4, and 43.2 ms for R = 4 and 8, respectively. Seven patients (mean age=41.5±20.7 years) undergoing clinical CMR were imaged in mid-ventricular short-axis and long-axis planes, following completion of the clinical examination using free breathing and electrocardiogram gating. The cine data sets were randomized and blinded for qualitative evaluation (image quality, artifact, noise; 1-5; lowest-highest) by a cardiologist and a radiologist. Statistical analysis was performed to compare the mean scores between the 4 groups (TGRAPPA-R4, TGRAPPA-R8, CS-PI-R4, CS-PI-R8) and between each pair of groups.


Figure 1 shows images of end-systolic frames in mid-ventricular short-axis and 2-chamber views. According to the Kruskal-Wallis test, the 4 groups were significantly different (p0.05): image quality, CS-PI-R4 vs. CS-PI-R8; artifact, CS-PI-R4 vs. CS-PI-R8; noise, TGRAPPA-R4 vs. CS-PI-R8, TGRAPPA-R4 vs. CS-PI-R4, CS-PI-R4 vs. CS-PI-R8. These preliminary results suggest that TGRAPPA can yield robust results at R=4, whereas CS-PI can yield robust results up to R=8.

thumbnailFigure 1. a) Short-axis view and b) long-axis view. For each view: (top row) TGRAPPA; (bottom row) joint CS-PI; (left column) R=4; (right column) R=8.


This study demonstrates the feasibility of performing highly-accelerated real-time cine MRI using a joint CS-PI technique. An 8-fold accelerated real-time cine MRI protocol can achieve spatial resolution of 2.5mm x 2.5mm and temporal resolution of 43.2 ms, with adequate image quality. This accelerated protocol may be useful for debilitated patients with reduced breath-hold capacity and/or arrhythmias for rapid left ventricular functional evaluation.

thumbnailFigure 2. Plots of mean scores: (left) image quality, (middle) artifact, and (right) noise. Pairs with no significant difference are noted with p>0.05.


  1. Kellman P, et al.:

    MRM. 2001. OpenURL

  2. Breuer FA, et al.:

    MRM. 2005. OpenURL

  3. Otazo R:

    MRM. 2010. OpenURL

  4. Lustig M, et al.:

    MRM. 2007. OpenURL