Comparison of 3 T and 7 T MRI clinical sequences for ankle imaging

doi:10.1016/j.ejrad.2011.05.023

European Journal of Radiology

Volume 81, Issue 8, August 2012, Pages 1846-1850

https://doi.org/10.1016/j.ejrad.2011.05.023 Get rights and content

Abstract

The purpose of this study was to compare 3 T and 7 T signal-to-noise and contrast-to noise ratios of clinical sequences for imaging of the ankles with optimized sequences and dedicated coils.

Ten healthy volunteers were examined consecutively on both systems with three clinical sequences: (1) 3D gradient-echo, T₁-weighted; (2) 2D fast spin-echo, PD-weighted; and (3) 2D spin-echo, T₁-weighted. SNR was calculated for six regions: cartilage; bone; muscle; synovial fluid; Achilles tendon; and Kager's fat-pad. CNR was obtained for cartilage/bone, cartilage/fluid, cartilage/muscle, and muscle/fat-pad, and compared by a one-way ANOVA test for repeated measures.

Mean SNR significantly increased at 7 T compared to 3 T for 3D GRE, and 2D TSE was 60.9% and 86.7%, respectively. In contrast, an average SNR decrease of almost 25% was observed in the 2D SE sequence. A CNR increase was observed in 2D TSE images, and in most 3D GRE images.

There was a substantial benefit from ultra high-field MR imaging of ankles with routine clinical sequences at 7 T compared to 3 T. Higher SNR and CNR at ultra-high field MR scanners may be useful in clinical practice for ankle imaging. However, carefully optimized protocols and dedicated extremity coils are necessary to obtain optimal results.

Introduction

Recently, the development of ultra-high-field MR scanners has grown rapidly. In theory, if the coil and the subject are considered equivalent, then the signal-to-noise ration (SNR) and contrast-to-noise ratio (CNR) would be almost linearly related to the static magnetic field B₀. SNR behavior, however, is apparently more complex, especially for human applications at 7 T (mostly due to susceptibility artifacts and a widened chemical shift) [1], [2].

In theory, the SNR between 3 T and 7 T should increase by a factor of 2.3. This is, however, not completely achievable in practical MRI. One of the main issues with ultra-high-field MR imaging is the specific absorption rate (SAR) limitation, which restricts a transfer of sequences from high-field scanners [3]. Recently developed parallel imaging techniques have great potential to overcome the SAR problem [4], [5]. To reduce artifacts, pulse properties and receiver bandwidth must be adjusted as well. Increased bandwidth should be employed to reduce the chemical shift artifacts between fat and water frequencies. Despite the intense developments in coil design for ultra-high-field imaging, there is still a lack of dedicated coils for musculoskeletal imaging, which are necessary to obtain a real benefit from higher field strengths. In this sense, multi-element phased array coils appear to be a promising solution [6].

MRI is widely used in ankle joint imaging for various anatomical regions, such as cartilage [7], tendons [8], or bones [9]. For most of these applications, high SNR is a prerequisite for successful diagnosis, or for advanced quantitative evaluation techniques. Higher field strengths via higher SNR and resolution may provide improvements in the biochemical assessment of cartilage, a better basis for trabecular bone segmentation, or in advanced imaging of tendons.

The aim of this study was to investigate the performance of clinical MR sequences in in vivo ankle imaging at ultra high-field (7 T), and compare it to the 3 T analogues. Different anatomical regions were compared for SNR and CNR. Ultra high-field MR imaging may substantially benefit from higher SNR and better spatial resolution.

Section snippets

Materials and methods

Ten healthy volunteers (mean age, 28.3 ± 6.4) were consecutively measured on 3 T (Tim Trio, Siemens Healthcare, Erlangen, Germany) and 7 T (Siemens Healthcare, Erlangen, Germany) MR scanners. The ethics commission of the Medical University of Vienna gave approval for this study, and written, informed consent was obtained from all volunteers before enrollment in the study. At 3 T, an eight-element dedicated knee coil (In Vivo, OR, USA) and at 7 T, a 28-channel dedicated knee coil (QED, USA) were used.

Results

The ICC revealed a very good agreement for nearly all measures for 3 T and 7 T (0.95 on average). The observed inter-observer variability was less than 8%. Overall mean SNR for 3 T was 23.64 ± 4.65, and for 7 T, 57.11 ± 13.7. SNR was significantly higher at 7 T compared to 3 T in most cases in the GRE and TSE sequences. The mean SNR increase for 3D GRE and 2D TSE was 60.9% and 86.7%, respectively. For 2D SE, an SNR decrease of almost 25% was observed. Non-significant differences were mostly found in SE

Discussion

In this study, the performance of standard clinical sequences was compared between 3 T and 7 T MRI. Imaging of the ankles at 7 T showed advantages compared with standard MR imaging protocols at 3 T. The study revealed a significantly better performance of clinical MR sequences in ankle joint imaging at 7 T compared to 3 T in two of three investigated sequences.

The greater benefit of increased SNR at 7 T has been previously demonstrated in brain imaging for detecting multiple sclerosis lesions [10],

Conclusion

A substantial benefit was shown for ultra high-field MR imaging of the ankles with routine clinical sequences at 7 T, compared to 3 T. Higher SNR and CNR at ultra high-field MR scanners may be useful in clinical practice for ankle imaging. However, carefully optimized protocols and dedicated extremity coils are necessary to obtain optimal results.

Acknowledgments

Funding for this study was provided by the Vienna Spots of Excellence des Wiener Wissenschafts-und Technologie-Fonds (WWTF) – Vienna Advanced Imaging Center – VIACLIC and the Slovak Scientific Grant Agency VEGA No. 2/0090/11.

References (20)

S.K. Pakin et al.
Ultra-high-field MRI of knee joint at 7.0 T: preliminary experience
Acad Radiol
(2006)
R.R. Regatte et al.
Ultra-high-field MRI of the musculoskeletal system at 7.0 T
J Magn Reson Imaging
(2007)
R. Krug et al.
Imaging of the musculoskeletal system in vivo using ultra-high field magnetic resonance at 7 T
Invest Radiol
(2009)
F. Schick et al.
MRI at high field: technical limits and clinical potential
Eur Radiol
(2005)
D.K. Sodickson et al.
Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays
Magn Reson Med
(1997)
K.P. Pruessmann et al.
SENSE: sensitivity encoding for fast MRI
Magn Reson Med
(1999)
P.B. Roemer et al.
The NMR phased array
Magn Reson Med
(1990)
S.A. Millington et al.
Quantitative and topographical evaluation of ankle articular cartilage using high resolution MRI
J Orthop Res
(2007)
M.E. Schweitzer et al.
MR imaging of disorders of the Achilles tendon
Am J Roentgenol
(2000)
T.M. Link et al.
In vivo high resolution MRI of the calcaneus: differences in trabecular structure in osteoporosis patients
J Bone Miner Res
(1998)

There are more references available in the full text version of this article.

Cited by (34)

Equine flexor tendon imaging part 2: Current status and future directions in advanced diagnostic imaging, with focus on the deep digital flexor tendon
2021, Veterinary Journal
Flexor tendon injuries are a common cause of lameness and early retirement in equine athletes. While ultrasonography is most frequently utilised, advanced diagnostic imaging modalities are becoming more widely available for detection and monitoring of flexor tendon lesions. Part two of this literature review details current experience with low- and high-field magnetic resonance imaging (MRI) and computed tomography (CT) for the diagnosis of equine flexor tendinopathy with a focus on the deep digital flexor tendon. Implications of the ‘magic angle’ artefact as well as injection techniques and the use of contrast media are discussed. Future developments in tendon imaging aim to gain enhanced structural information about the tendon architecture with the prospect to prevent injury. Techniques as described for the assessment of the human Achilles tendon including ultra-high field MRI and positron emission tomography are highlighted.
Magnetic resonance imaging for three-dimensional printing of the bony orbit: is clinical use imminent?
2020, International Journal of Oral and Maxillofacial Surgery
Citation Excerpt :
In the future, it would be reasonable to expect a greater number of patients presenting to maxillofacial surgeons with orbital fractures who have not had CT imaging. Recent advancements in MRI such as 7T may increase the accuracy of MRI for orbital fractures, owing to increased definition of the soft and hard tissue, thus increasing the potential to detect even slight disruptions to the normal anatomy29–31. This may also lend itself to the potential for automated segmentation of MR images, particularly in combination with machine learning.
The aim of this study was to examine the accuracy of three dimensionally (3D) printed models of the bony orbit derived from magnetic resonance imaging (MRI) for the purpose of preoperative plate bending in the setting of orbital blowout fracture. Retrospective computed tomography (CT) and MRI data from patients with suspected orbital fractures were used. Virtual models were manually generated and analysed for spatial accuracy of the fracture margins. 3D-printed models were produced and orbital fan plates bent by a single operator. The plates were then digitized and analysed for spatial discrepancy using reverse engineering software. Seven orbital blowout fractures were evident in six orbits. Analysis of the virtual models revealed high congruence between blowout fracture margins on CT and MRI (n = 7, average deviation 0.85 mm). Three zygomaticomaxillary complex fractures were seen, for which MRI did not demonstrate the same accuracy. For plates bent to the 3D-printed models of blowout fractures (n = 6), no significant difference was found between those bent to CT versus those bent to MRI when compared for average surface and average border deviation (Wilcoxon signed rank test). Orbital blowout fractures can be defined on MRI with clinically acceptable accuracy. 3D printing of orbital biomodels from MRI for bending reconstructive plates is an acceptable and accurate technique.
A new RF transmit coil for foot and ankle imaging at 7T MRI
2018, Magnetic Resonance Imaging
Citation Excerpt :
The lack of Tx body coil for 7T MRI forces the development of transmit coils dedicated for certain parts of the body. Due to the absence of commercial RF coil specific for foot/ankle at 7T, most sites use knee coils [7,18–22] or head coils [23–26] for imaging the foot/ankle. This situation is not optimal for several reasons.
A four-channel Tic-Tac-Toe (TTT) transmit RF coil was designed and constructed for foot and ankle imaging at 7T MRI. Numerical simulations using an in-house developed FDTD package and experimental analyses using a homogenous phantom show an excellent agreement in terms of B₁⁺ field distribution and s-parameters. Simulations performed on an anatomically detailed human lower leg model demonstrated an B₁⁺ field distribution with a coefficient of variation (CV) of 23.9%/15.6%/28.8% and average B₁⁺ of 0.33 μT/0.56 μT/0.43 μT for 1 W input power (i.e., 0.25 W per channel) in the ankle/calcaneus/mid foot respectively. In-vivo B₁⁺ mapping shows an average B₁⁺ of 0.29 μT over the entire foot/ankle. This newly developed RF coil also presents acceptable levels of average SAR (0.07 W/kg for 10 g per 1 W of input power) and peak SAR (0.34 W/kg for 10 g per 1 W of input power) over the whole lower leg. Preliminary in-vivo images in the foot/ankle were acquired using the T2-DESS MRI sequence without the use of a dedicated receive-only array.
Delayed gadolinium-enhanced MRI of the fibrocartilage disc of the temporomandibular joint - a feasibility study
2014, Magnetic Resonance Imaging
Citation Excerpt :
The dGEMRIC index, i.e., the T1 relaxation time following (Gd-DTPA)2- administration (T1(Gd)), is an indirect measure of the glycosaminoglycan (GAG) concentration of cartilage tissue [18–20]. At field-strengths of 3 T, the biochemical MRI measurement of smaller joint cartilage, such as the ankle joint or lumbar facets, becomes possible in satisfactory image resolution and clinically reasonable measurement time [21–23]. Recently, these biochemical techniques were adapted to fibrocartilaginous tissues, such as the menisci [24,25], where, similar to the fibrocartilage structure of the TMJ disc, GAGs are less abundant compared to hyaline cartilage [2,26].
To 1) test the feasibility of delayed Gadolinium-Enhanced Magnetic Resonance Imaging of Cartilage (dGEMRIC) at 3 T in the temporomandibular joint (TMJ) and 2) to determine the optimal delay for measurements of the TMJ disc after i.v. contrast agent (CA) administration.
MRI of the right and left TMJ of six asymptomatic volunteers was performed at 3 T using a dedicated coil. 2D inversion recovery (2D-IR) sequences were performed at 4 time points covering 120 minutes and 3D gradient-echo (3D GRE) dual flip-angle sequences were performed at 14 time points covering 130 minutes after the administration of 0.2 mmol/kg of Gd-diethylenetriamine pentaacetic acid ion (Gd-DTPA)^2-, i.e., 0.4 mL of Magnevist™ per kg body weight. Pair-wise tests were used to assess differences between pre-and post-contrast T1 values.
2D-IR sequences showed a statistically significant drop (p < 0.001) in T1 values after i.v. CA administration. The T1 drop of 50% was reached 60 minutes after bolus injection in the TMJ disc. The 3D GRE dual flip-angle sequences confirmed these results and show plateau of T1 after 60 minutes.
T1(Gd) maps calculated from dGEMRIC data allow in vivo assessment of the fibrocartilage disc of the TMJ. The recommended measurement time for dGEMRIC in the TMJ after i.v. CA administration is from 60 to 120 minutes.
Optimizing MRI of small joints and extremities
2014, Clinical Radiology
Citation Excerpt :
Groups investigating 7 T imaging of the wrist and ankle have demonstrated variable increases in the SNR between different tissue types. This has been attributed to the parameters and coils used, which are still in the early stages of development.7–9 For a given MRI machine, the choice of receiver coil is the most effective way to improve SNR, without cost to other imaging parameters.
Obtaining optimal images of small joints using magnetic resonance imaging (MRI) can be technically challenging. The aim of this review is to outline the practical aspects of MRI of small joints, with reference to the underlying physical principles. Although the most important contribution to successful imaging of small joints comes from the magnet field strength and design of the receiver coil, there are a number of factors to balance including the signal-to-noise ratio, image resolution, and acquisition times. We discuss strategies to minimize artefacts from movement, inhomogeneity, chemical shift, and fat suppression. As with all MRI, each strategy comes at a price, but the benefits and costs of each approach can be fine-tuned to each combination of joint, receiver coil, and MRI machine.
7 T Musculoskeletal MRI: Fundamentals and Clinical Implementation
2023, Investigative Radiology

View all citing articles on Scopus

View full text

Comparison of 3 T and 7 T MRI clinical sequences for ankle imaging

Abstract

Introduction

Section snippets

Materials and methods

Results

Discussion

Conclusion

Acknowledgments

Acad Radiol

Ultra-high-field MRI of the musculoskeletal system at 7.0 T

J Magn Reson Imaging

Imaging of the musculoskeletal system in vivo using ultra-high field magnetic resonance at 7 T

Invest Radiol

MRI at high field: technical limits and clinical potential

Eur Radiol

Simultaneous acquisition of spatial harmonics (SMASH): fast imaging with radiofrequency coil arrays

Magn Reson Med

SENSE: sensitivity encoding for fast MRI

Magn Reson Med

The NMR phased array

Magn Reson Med

Quantitative and topographical evaluation of ankle articular cartilage using high resolution MRI

J Orthop Res

MR imaging of disorders of the Achilles tendon

Am J Roentgenol

In vivo high resolution MRI of the calcaneus: differences in trabecular structure in osteoporosis patients

J Bone Miner Res