Accelerated cardiovascular magnetic resonance of the mouse heart using self-gated parallel imaging strategies does not compromise accuracy of structural and functional measures
© Ratering et al; licensee BioMed Central Ltd. 2010
Received: 7 February 2010
Accepted: 21 July 2010
Published: 21 July 2010
Self-gated dynamic cardiovascular magnetic resonance (CMR) enables non-invasive visualization of the heart and accurate assessment of cardiac function in mouse models of human disease. However, self-gated CMR requires the acquisition of large datasets to ensure accurate and artifact-free reconstruction of cardiac cines and is therefore hampered by long acquisition times putting high demands on the physiological stability of the animal. For this reason, we evaluated the feasibility of accelerating the data collection using the parallel imaging technique SENSE with respect to both anatomical definition and cardiac function quantification.
Findings obtained from accelerated data sets were compared to fully sampled reference data. Our results revealed only minor differences in image quality of short- and long-axis cardiac cines: small anatomical structures (papillary muscles and the aortic valve) and left-ventricular (LV) remodeling after myocardial infarction (MI) were accurately detected even for 3-fold accelerated data acquisition using a four-element phased array coil. Quantitative analysis of LV cardiac function (end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), ejection fraction (EF) and LV mass) in healthy and infarcted animals revealed no substantial deviations from reference (fully sampled) data for all investigated acceleration factors with deviations ranging from 2% to 6% in healthy animals and from 2% to 8% in infarcted mice for the highest acceleration factor of 3.0. CNR calculations performed between LV myocardial wall and LV cavity revealed a maximum CNR decrease of 50% for the 3-fold accelerated data acquisition when compared to the fully-sampled acquisition.
We have demonstrated the feasibility of accelerated self-gated retrospective CMR in mice using the parallel imaging technique SENSE. The proposed method led to considerably reduced acquisition times, while preserving high spatial resolution at sufficiently high CNR. The accuracy of measurements of both structural and functional parameters of the mouse heart was not compromised by the application of the proposed accelerated data acquisition method.
In view of increasing availability of genetically engineered strains, mouse models of human disease have become important tools for the study of pathophysiological mechanisms and the evaluation of novel therapies. A remarkable number of such transgenic models have been developed mimicking aspects of cardiovascular pathologies such as heart failure and myocardial infarction (MI). Phenotyping of disease models commonly includes investigation of the morphology and function of the heart, preferentially using non-invasive imaging methods such as cardiovascular magnetic resonance (CMR), computed tomography (CT)  and echocardiography . In particular CMR has evolved as an accurate modality for the characterization of cardiac pathophysiology due to the inherent high soft-tissue contrast and the high spatial resolution of the method. On the other hand, dynamic CMR studies of the mouse heart face several challenges. Due to the small dimensions (the volume is three orders of magnitude smaller than the human heart) and the high heart and breathing rates of mice (heart rate: 400 - 550 beats per minute, breathing rate: 60 - 150 breaths per minute), imaging methods developed for humans can not be translated to CMR of small animals in a straightforward manner.
Cardiac MR images can be acquired following two different strategies: either using the prospective approach [3–9], i.e. MR data acquisition is directly triggered by the ECG signal, or using the retrospective approach [10, 11], i.e., data is continuously collected during the experiment and subsequently reconstructed by using the stored respiratory and ECG information to assign the data to respiratory and/or cardiac phases.
In recent reports, self-gated cardiac imaging methods have been demonstrated to be a valuable tool for cardiac MR imaging in small rodents [12, 13], since neither ECG leads nor respiratory motion sensors have to be utilized. The advantages of using self-gated retrospective CMR in small animal imaging are counterbalanced by long data acquisition times. Indeed, the acquisition of large datasets is required to ensure accurate and artifact-free reconstruction of cardiac movies. Long scan durations are associated with disadvantages, the most trivial one being low throughput and hence lack of efficiency. More fundamental are the high demands on the physiological/metabolic stability of the animal under investigation. Stable conditions must be maintained throughout the experiment, a particular challenge when studying diseased animals.
Accelerating self-gated cardiac imaging is therefore of high interest as demonstrated in clinical studies . So-called parallel imaging techniques such as SENSitivity Encoding (SENSE)  represent an efficient strategy to reduce scan time, however, at the expense of decreased signal-to-noise (SNR) ratios. This is a particular concern in small animal imaging, which is inherently compromised by low SNR values due to the small voxel dimensions involved. In view of the high demands regarding sensitivity in rodent CMR and the inherent losses of SNR associated with accelerated data acquisition, the application of parallel imaging techniques in small animal CMR remains a challenge.
In this work the feasibility of accelerating self-gated cardiac imaging for assessing ventricular function in mice has been evaluated. In a first step, SENSE accelerated self-gated CMR has been adapted and optimized in order to assess morphological and functional parameters of the normal mouse heart. Subsequently, the method has been applied to self-gated CMR in mice subjected to experimental MI one week prior to the examination. The sensitivity of non-accelerated (fully sampled) self-gated retrospective CMR data in detecting structural and functional alterations due to MI was compared with that of accelerated acquisitions. Acceleration factors of up to 3.0 were investigated, resulting in a substantial decrease of the data acquisition time (from typically 20 to 6 min). For detailed comparison, contrast-to-noise ratios (CNRs) of myocardial wall as well as functional heart parameters such as left ventricular (LV) end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV), ejection fraction (EF) and LV mass have been calculated and compared between datasets derived from non-accelerated (acquisition time: 18.7 min) and accelerated scans (acquisition time: 6.2 min for three-fold acceleration).
Materials and methods
Animal Model and Preparation
All animal experiments were carried out in strict adherence to the Swiss law for animal protection. CMR acquisitions were performed in eight healthy and six infarcted female C57Bl/6J mice showing average body weights of 24.0 ± 2.1 g. MI was induced by permanent ligation of the left anterior descending coronary artery. The surgical intervention was performed as described in detail in literature [16, 17]. The infarcted animals were measured one week after the surgical intervention.
For CMR experiments, mice were anesthetized with an initial dose of 3% isoflurane (Abbott, Cham, Switzerland) in air:O2 (4:1) and positioned on a cradle made of PMMA comprising a built-in warm water circuit. Animals were freely breathing and carefully positioned with the heart resting on the center of the RF receive coil. Body temperature was measured using a rectal temperature probe (MLT415, ADInstruments, Spechbach, Germany). Special care was taken to maintain body temperature at 36 ± 0.5°C. During the experiments anesthesia was maintained using a reduced level of isoflurane (1.6%). Mean heart and mean respiratory rates of the animals were estimated retrospectively form the acquired navigator signal yielding 498 ± 46 beats/min and 85 ± 16 breaths/min, respectively.
All experiments were performed using a Bruker Pharmascan 47/16 (Bruker BioSpin MRI, Ettlingen, Germany) small animal MR system operating at 200 MHz with a gradient system capable of switching 350 mT/m in 200 μs. A volume resonator (75.4 mm inner diameter, active length = 70 mm) operating in quadrature mode was used for excitation and a four element (2 × 2) phased array surface coil (outer dimensions of one coil element: 12 × 16 mm2, total outer dimensions: 26 × 21 mm2) for signal reception. The coil elements were bent to a cylindrical surface (radius = 10 mm) with the axis parallel to the main magnetic field direction. The proximity of the four coil elements allowed large acceleration factors (up to three-fold acceleration) without considerably degrading image quality as each part of the imaging slice provided enough signal in all four coil elements.
Scout images of the heart anatomy were acquired for accurate planning of the subsequent cine cardiac scans. Black-blood short-axis images covering the entire left ventricle were acquired using the self-gating technique IntraGate (Bruker BioSpin MRI, Ettlingen, Germany), which is based on a fast low-angle shot (FLASH)  multislice sequence with an extra navigator echo. The following parameters were used for data acquisition: multislice gradient-echo sequence, field-of-view (FOV) = 25 × 25 mm2, matrix dimension = 154 × 154, spatial resolution = 162 × 162 μm2, 6 to 7 contiguous slices of 1.0 mm thickness, pulse angle = 10°, echo/repetition time (TE/TR) = 1.6/5.2 ms, number of repetitions (NR) = 200, total acquisition time = 18.7 min. Inflowing blood was saturated in a slice of 5.0 mm thickness parallel to the imaging stack with a saturation pulse angle of 60° and carefully placed at the inflow tract of the ventricles with a gap of 0.5 mm between saturation slice and first imaging slice. In addition, the navigator signal was derived from the saturation slice prior to the excitation of the cardiac imaging slices. Cardiac data were collected either in a non-accelerated manner (full k-space sampling) or accelerated in phase encoding direction (reduced k-space sampling). Acceleration factors (R) used were 1.5, 2.0, 2.5 and 3.0, resulting in total acquisition times of 12.4 min, 9.4 min, 7.5 min and 6.2 min, respectively.
In addition, the feasibility of accelerated data collection was investigated for cardiac cine images acquired in cardiac long axis orientation using a self-gated bright-blood sequence with the following parameters: FOV = 24 × 24 mm2, matrix dimension = 288 × 288, spatial resolution = 83 × 83 μm2, slice thickness = 1.0 mm, pulse angle = 15°, echo/repetition time (TE/TR) = 2.7/10 ms, number of repetitions (NR) = 300. The navigator slice was positioned at the same location as the imaging slice with a slice thickness of 1 mm. Heart data were acquired for acceleration factors of 1.0, 2.0 and 3.0, corresponding to scan durations of 14.4 min, 7.2 min and 4.8 min, respectively.
Cardiac cine images were reconstructed by using the collected navigator information within the IntraGate software to assign the acquired k-space lines to ten cardiac frames and one respiratory frame, retrospectively.
where MEAN(S Noise ) and SD(S Noise ) represent mean and standard deviation of signal originating from a region-of-interest (ROI) placed in an image region exhibiting pure noise.
As the volume resonator could only be operated in transmission mode, the magnitude of the reference image was calculated as the sum-of-squares (SOS) of the single coil images and the corresponding phase according to the method proposed by de Zwart et al. . The noise covariance matrix of the coil elements was estimated from a separate scan and included in the reconstruction algorithm as proposed by Pruessmann et al. . The reconstruction algorithm was performed on a separate workstation using in-house software written in IDL (RSI, Boulder, USA).
Validation and optimization of the proposed accelerated CMR acquisition and reconstruction method was first performed in the healthy animals: CNR analyses as well as the calculation of EDV, ESV, SV and EF were carried out in order to investigate the possible effects of accelerated data collection on image quality and on the accuracy of image derived cardiac parameters. Subsequently, the optimized acquisition and reconstruction parameters were utilized to investigate the sensitivity of the proposed accelerated CMR method in assessing myocardial remodeling induced by MI in six diseased animals.
Cardiac functional parameters were estimated from black-blood short-axis images of the left ventricle. In order to segment the left ventricular cavity a ROI was manually selected covering both ventricular cavity and wall. Subsequently, left ventricular volume and cavity were discriminated using a threshold based algorithm providing a binary image for each slice. The threshold was defined as 0.6 • MEAN(S Left Ventricle ), where MEAN(S Left Ventricle ) represents the average signal intensity of the whole left ventricle. The threshold value for intensity based image segmentation was optimized by comparing the LV volumes obtained from the MR data with references volumes from literature [6, 7, 20]. The best agreement was obtained for a factor of 0.6. Voxels with signal intensity higher than the threshold were attributed to the LV wall, voxels with signal intensity lower than the threshold to the LV cavity. Segmentation of cardiac images and calculation of EDV and ESV was performed using in-house software written in IDL.
where S Myocardial Wall , S Ventricular Cavity , SD(S Myocardial Wall ) and SD(S Ventricular Cavity ) are defined as the signal of the myocardial wall, the signal of the ventricular cavity and the standard deviation of the myocardial wall and ventricular cavity signal, respectively. All CNR analyses of this study were performed for the accelerated self-gated retrospective black-blood measurements using in-house software written in IDL. CNR calculations were performed for all acceleration factors under investigation according to the threshold based segmentation algorithm described previously.
where P LV; R,i is defined as the LV parameter for acceleration factor R (R = 1.5, 2.0, 2.5, 3.0) and P LV; R = 1.0,i as the LV parameter for acceleration factor 1.0 for each animal i. Calculations according to Eq.  were performed for the LV functional parameters EDV, ESV, SV, EF and LV mass.
Finally, in order to assess reproducibility of our proposed method, inter- and intra-observer variability measurements were performed by two independent observers as well as twice by one observer for all investigated cardiac functional parameters for R = 1.0 and R = 3.0.
Impact of accelerated CMR on the anatomical definition of the mouse heart
Impact of accelerated CMR on quantitative functional ventricular parameters
Quantitative assessment of LV function in eight healthy animals: EDV, ESV, SV and EF parameters were derived for all investigated acceleration factors R.
R = 1.0
R = 1.5
R = 2.0
R = 2.5
R = 3.0
44.2 ± 2.0
44.3 ± 1.8
45.2 ± 2.2
45.0 ± 1.9
44.2 ± 2.5
12.1 ± 0.8
12.2 ± 0.9
12.4 ± 1.0
12.4 ± 1.1
12.2 ± 1.3
32.2 ± 1.5
32.1 ± 1.2
32.6 ± 1.4
32.6 ± 1.7
32.0 ± 1.6
72.8 ± 1.1
72.5 ± 1.3
72.6 ± 1.4
72.5 ± 2.2
72.5 ± 1.9
Finally, the mean deviation of the various functional parameters (ΔP LV ) from the reference values obtained from the R = 1.0 reference datasets was analyzed for all investigated acceleration factors according to Eq. . Using the absolute value of differences, i.e. Eq. , allowed to avoid possible averaging effects when calculating the mean values of the different LV parameters. The results are summarized in Figure 5c: the maximum mean deviation ΔP LV of 6% was found for the ESV for an acceleration factor of R = 3.0. All other mean deviations estimated for EDV, SV and EF were found to be < 5% demonstrating that acceleration of data acquisition did not significantly compromise the accuracy of morphometric measurements.
Inter- and intra-observer variability assessment in eight healthy animals for all investigated LV parameters: EDV, ESV, SV, EF and mass.
Inter-observer variability [%]
Intra-observer variability [%]
R = 1.0
R = 3.0
R = 1.0
R = 3.0
2.9 ± 1.8
3.4 ± 1.7
2.3 ± 1.9
2.9 ± 2.5
3.3 ± 3.9
4.1 ± 3.1
3.1 ± 2.4
3.9 ± 2.9
4.5 ± 2.2
5.3 ± 1.9
3.9 ± 2.5
4.9 ± 2.6
3.3 ± 1.6
3.9 ± 2.4
2.8 ± 1.9
3.4 ± 2.4
4.3 ± 2.6
4.4 ± 4.1
4.1 ± 2.1
4.8 ± 3.9
Sensitivity of accelerated CMR in assessing myocardial remodeling after MI
Quantitative assessment of LV function in six animals subjected to MI: EDV, ESV, SV and EF parameters were derived for all investigated acceleration factors R.
R = 1.0
R = 1.5
R = 2.0
R = 2.5
R = 3.0
86.4 ± 3.8
85.9 ± 4.7
84.2 ± 7.3
87.4 ± 4.9
88.2 ± 5.5
68.1 ± 3.9
68.1 ± 5.4
66.9 ± 6.9
69.1 ± 5.4
68.9 ± 5.5
17.9 ± 1.2
17.8 ± 1.1
17.4 ± 1.5
18.3 ± 0.9
19.0 ± 2.5
22.8 ± 1.6
22.9 ± 2.2
22.9 ± 2.1
23.1 ± 1.9
24.0 ± 2.8
In summary, only minor deviations from the reference values (LV parameters obtained for R = 1.0) could be found for all calculated cardiac functional parameters.
Self-gated CMR has been shown to yield accurate structural and functional measures of the mouse heart [12, 13]. Yet, the approach suffers from long acquisition times in order to acquire a large enough data basis for reconstructing the different phases of the cardiac cycle. Our study shows that the self-gated CMR data acquisition scheme can be successfully accelerated yielding CNR values that are sufficient for an accurate assessment of the dynamic LV morphology in mice. Significant shortening of acquisition times by factors between 1.5 and 3.0 have been achieved using the parallel CMR approach SENSE, while still preserving good image quality. Indeed, the results of this study demonstrate that 3-fold acceleration of multi-slice CMR of the mouse heart can be successfully achieved (R = 3.0) in combination with a four-element phased array coil. This translates into a scan time reduction from 18.7 min (R = 1.0) to 6.2 min (R = 3.0) for the multi-slice black-blood short-axis view and from 14.4 min (R = 1.0) to 4.8 min (R = 3.0) for the long-axis bright-blood view of the mouse heart. Similar scan time reductions have been reported by J.E. Schneider and coworkers , which have investigated the effects of TGRAPPA  reconstruction on cardiac function assessment using a four-element (4 × 1) RF coil array. However, in comparison to our study, measurements have been carried out in rats at a field strength of 9.4 T using an ECG-triggered and respiratory gated sequence. In addition, the accelerated cardiac data sets were simulated offline form the acquired fully sampled acquisitions. It is shown, that measurement times could be decreased from 11 min (R = 1.0) to approximately 3 min for an acceleration factor of R = 4.0 without severely compromising image quality and cardiac function assessment.
Even though CNR decreases due to inherent sensitivity losses of SENSE reconstructed data, good quality black-blood and bright-blood images of the mouse heart could be obtained for all acceleration factors. Indeed, despite reductions in CNR Wall-Cavity calculated between LV myocardial wall and LV cavity (45 - 50% for R = 3.0, Figure 4), high definition short- and long-axis views of the mouse heart have been obtained in both healthy and diseased animals. This is demonstrated by the excellent definition of the aortic valve in the bright-blood images (Figure 3), the dynamics of which can be monitored throughout the cardiac cycle. Similarly, for all R values contrast between LV myocardial wall and cavity in black-blood images was sufficient to allow for unambiguous delineation of epi- and endocardial borders, an important prerequisite for accurate determination of functional cardiac parameters. Small anatomical structures, such as the papillary muscles could be depicted with excellent anatomical detail even for an acceleration factor of R = 3.0 (Figures 1, 2 and 3). Furthermore, the proposed accelerated acquisition method was sensitive enough to accurately detect myocardial remodeling after MI, such as LV enlargement and LV wall thickening/thinning, even for an acceleration factor of R = 3.0 (Figures 3, 6 and 7).
LV functional parameters derived from the acquired data of both healthy and infarcted animals were found to be comparable with values already reported in literature [1, 6, 7, 11, 20]. The good agreement between non-accelerated and accelerated acquisitions is documented by the correlation of cross-sectional LV cavity areas (Figure 5 a, b) and the small mean deviations of LV functional parameters (ΔP LV ) derived from accelerated acquisitions from values obtained from fully-sampled data (Figures 5c and 8). Thus, the quantitative analysis of cardiac function revealed that accelerating data acquisition did not substantially reduce the precision in assessing LV functional parameters in both healthy and diseased animals. In addition, reproducibility of the proposed accelerated data acquisition method in assessing LV function of the mouse heart was demonstrated from findings obtained from inter- and intra-observer variability measurements (Table 2).
Increasing spatial in-plane resolution for 2D imaging or performing four-dimensional CMR (4D-CMR, 3 spatial and 1 temporal dimension) allowing for high spatial resolution in all three dimensions would have beneficial effects on the depiction of the cardiac anatomy and the derivation of cardiac functional parameters by minimizing partial volume effects. On the other hand, increasing spatial resolution has the drawback of increased scan durations, making the application of the proposed accelerated acquisition method even more attractive. In plane spatial resolution in the current study was 83 × 83 μm2 for bright-blood and 162 × 162 μm2 for black-blood images. While resolution can be improved in principle, there are practical limitations. Reducing the voxel size will decrease SNR and consequently CNR, making unambiguous identification of tissue borders difficult. A second potentially limiting factor is the size of the acquired four-dimensional dataset, which will translate into long processing times.
The sensitivity issue can be addressed by applying the SENSE based accelerated data acquisition method to 4D self-gated retrospective CMR. Furthermore, the spatial encoding efficiency of a receiver array is enhanced for 3D imaging since acceleration can be done in both phase encoding directions . Finally, higher acceleration factors and/or better image quality of SENSE reconstructed cardiac images might be achievable by increasing the number of coil elements of the receiver array or by optimizing the coil geometry and hence reducing local noise enhancement in the region of the heart .
In conclusion, we have demonstrated the feasibility of accelerating self-gated retrospective CMR in mice using the parallel imaging technique SENSE in combination with a four-element phased array surface coil for signal detection. The accuracy of the accelerated acquisition allows identification of small anatomical heart structures such as the papillary muscles and the aortic valve as well as a reliable quantification of left ventricular function. Accelerating data collection using SENSE led to considerably reduced acquisition times (up to factor 3.0), while preserving high spatial resolution at sufficiently high CNR, a prerequisite for accurate calculation of cardiac functional parameters. Finally, quantitative analysis of cardiac function and assessment of myocardial remodeling in infarcted animals was successfully performed demonstrating the feasibility of the developed method to follow disease progression and therapy.
We would like to acknowledge financial support by the National Competence Center for Research (NCCR), Neural Plasticity and Repair and the Swiss National Science Foundation.
- Nahrendorf M, Badea C, Hedlund LW, Figueiredo JL, Sosnovik DE, Johnson GA, Weissleder R: High-resolution imaging of murine myocardial infarction with delayed-enhancement cine micro-CT. Am J Physiol Heart Circ Physiol. 2007, 292: H3172-3178. 10.1152/ajpheart.01307.2006.PubMed CentralView ArticlePubMedGoogle Scholar
- Rottman JN, Ni G, Brown M: Echocardiographic evaluation of ventricular function in mice. Echocardiography. 2007, 24: 83-89. 10.1111/j.1540-8175.2006.00356.x.View ArticlePubMedGoogle Scholar
- Bucholz E, Ghaghada K, Qi Y, Mukundan S, Johnson GA: Four-dimensional MR microscopy of the mouse heart using radial acquisition and liposomal gadolinium contrast agent. Magn Reson Med. 2008, 60: 111-118. 10.1002/mrm.21618.PubMed CentralView ArticlePubMedGoogle Scholar
- Feintuch A, Zhu Y, Bishop J, Davidson L, Dazai J, Bruneau BG, Henkelman RM: 4D cardiac MRI in the mouse. NMR Biomed. 2007, 20: 360-365. 10.1002/nbm.1164.View ArticlePubMedGoogle Scholar
- Rose SE, Wilson SJ, Zelaya FO, Crozier S, Doddrell DM: High resolution high field rodent cardiac imaging with flow enhancement suppression. Magn Reson Imaging. 1994, 12: 1183-1190. 10.1016/0730-725X(94)90084-5.View ArticlePubMedGoogle Scholar
- Ruff J, Wiesmann F, Hiller KH, Voll S, von Kienlin M, Bauer WR, Rommel E, Neubauer S, Haase A: Magnetic resonance microimaging for noninvasive quantification of myocardial function and mass in the mouse. Magn Reson Med. 1998, 40: 43-48. 10.1002/mrm.1910400106.View ArticlePubMedGoogle Scholar
- Schneider JE, Cassidy PJ, Lygate C, Tyler DJ, Wiesmann F, Grieve SM, Hulbert K, Clarke K, Neubauer S: Fast, high-resolution in vivo cine magnetic resonance imaging in normal and failing mouse hearts on a vertical 11.7 T system. J Magn Reson Imaging. 2003, 18: 691-701. 10.1002/jmri.10411.View ArticlePubMedGoogle Scholar
- Slawson SE, Roman BB, Williams DS, Koretsky AP: Cardiac MRI of the normal and hypertrophied mouse heart. Magn Reson Med. 1998, 39: 980-987. 10.1002/mrm.1910390616.View ArticlePubMedGoogle Scholar
- Wiesmann F, Ruff J, Haase A: High-resolution MR imaging in mice. Magma. 1998, 6: 186-188. 10.1007/BF02660956.View ArticlePubMedGoogle Scholar
- Bishop J, Feintuch A, Bock NA, Nieman B, Dazai J, Davidson L, Henkelman RM: Retrospective gating for mouse cardiac MRI. Magn Reson Med. 2006, 55: 472-477. 10.1002/mrm.20794.View ArticlePubMedGoogle Scholar
- Miraux S, Calmettes G, Massot P, Lefrancois W, Parzy E, Muller B, Arsac LM, Deschodt-Arsac V, Franconi JM, Diolez P, Thiaudiere E: 4D retrospective black blood trueFISP imaging of mouse heart. Magn Reson Med. 2009, 62: 1099-1105. 10.1002/mrm.22139.View ArticlePubMedGoogle Scholar
- Heijman E, de Graaf W, Niessen P, Nauerth A, van Eys G, de Graaf L, Nicolay K, Strijkers GJ: Comparison between prospective and retrospective triggering for mouse cardiac MRI. NMR Biomed. 2007, 20: 439-447. 10.1002/nbm.1110.View ArticlePubMedGoogle Scholar
- Hiba B, Richard N, Thibault H, Janier M: Cardiac and respiratory self-gated cine MRI in the mouse: comparison between radial and rectilinear techniques at 7T. Magn Reson Med. 2007, 58: 745-753. 10.1002/mrm.21355.View ArticlePubMedGoogle Scholar
- Uribe S, Muthurangu V, Boubertakh R, Schaeffter T, Razavi R, Hill DL, Hansen MS: Whole-heart cine MRI using real-time respiratory self-gating. Magn Reson Med. 2007, 57: 606-613. 10.1002/mrm.21156.View ArticlePubMedGoogle Scholar
- Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P: SENSE: sensitivity encoding for fast MRI. Magn Reson Med. 1999, 42: 952-962. 10.1002/(SICI)1522-2594(199911)42:5<952::AID-MRM16>3.0.CO;2-S.View ArticlePubMedGoogle Scholar
- Engberding N, Spiekermann S, Schaefer A, Heineke A, Wiencke A, Muller M, Fuchs M, Hilfiker-Kleiner D, Hornig B, Drexler H, Landmesser U: Allopurinol attenuates left ventricular remodeling and dysfunction after experimental myocardial infarction: a new action for an old drug?. Circulation. 2004, 110: 2175-2179. 10.1161/01.CIR.0000144303.24894.1C.View ArticlePubMedGoogle Scholar
- Landmesser U, Engberding N, Bahlmann FH, Schaefer A, Wiencke A, Heineke A, Spiekermann S, Hilfiker-Kleiner D, Templin C, Kotlarz D, Mueller M, Fuchs M, Hornig B, Haller H, Drexler H: Statin-induced improvement of endothelial progenitor cell mobilization, myocardial neovascularization, left ventricular function, and survival after experimental myocardial infarction requires endothelial nitric oxide synthase. Circulation. 2004, 110: 1933-1939. 10.1161/01.CIR.0000143232.67642.7A.View ArticlePubMedGoogle Scholar
- Frahm J, Haase A, Matthaei D: Rapid NMR imaging of dynamic processes using the FLASH technique. Magn Reson Med. 1986, 3: 321-327. 10.1002/mrm.1910030217.View ArticlePubMedGoogle Scholar
- de Zwart JA, van Gelderen P, Kellman P, Duyn JH: Application of sensitivity-encoded echo-planar imaging for blood oxygen level-dependent functional brain imaging. Magn Reson Med. 2002, 48: 1011-1020. 10.1002/mrm.10303.View ArticlePubMedGoogle Scholar
- Winter EM, Grauss RW, Atsma DE, Hogers B, Poelmann RE, van der Geest RJ, Tschope C, Schalij MJ, Gittenberger-de Groot AC, Steendijk P: Left ventricular function in the post-infarct failing mouse heart by magnetic resonance imaging and conductance catheter: a comparative analysis. Acta Physiol (Oxf). 2008, 194: 111-122. 10.1111/j.1748-1716.2008.01863.x.View ArticleGoogle Scholar
- Franco F, Dubois SK, Peshock RM, Shohet RV: Magnetic resonance imaging accurately estimates LV mass in a transgenic mouse model of cardiac hypertrophy. Am J Physiol. 1998, 274: H679-683.PubMedGoogle Scholar
- Roemer PB, Edelstein WA, Hayes CE, Souza SP, Mueller OM: The NMR phased array. Magn Reson Med. 1990, 16: 192-225. 10.1002/mrm.1910160203.View ArticlePubMedGoogle Scholar
- Schneider JE, Lanz T, Barnes H, Medway D, Stork LA, Lygate CA, Smart S, Griswold MA, Neubauer S: Ultra-fast and accurate assessment of cardiac function in rats using accelerated MRI at 9.4 Tesla. Magn Reson Med. 2008, 59: 636-641. 10.1002/mrm.21491.View ArticlePubMedGoogle Scholar
- Breuer FA, Kellman P, Griswold MA, Jakob PM: Dynamic autocalibrated parallel imaging using temporal GRAPPA (TGRAPPA). Magn Reson Med. 2005, 53: 981-985. 10.1002/mrm.20430.View ArticlePubMedGoogle Scholar
- Weiger M, Pruessmann KP, Boesiger P: 2D SENSE for faster 3D MRI. Magma. 2002, 14: 10-19. 10.1007/BF02668182.View ArticlePubMedGoogle Scholar
- Weiger M, Pruessmann KP, Leussler C, Roschmann P, Boesiger P: Specific coil design for SENSE: a six-element cardiac array. Magn Reson Med. 2001, 45: 495-504. 10.1002/1522-2594(200103)45:3<495::AID-MRM1065>3.0.CO;2-V.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.