Importance of imaging plane for magnetic resonance imaging of the normal left ventricle.

Visualization of the left ventricle by magnetic resonance imaging (MRI) is largely a function of imaging planes. The best depiction of the myocardium is possible when the imaging plane is perpendicular to the region of interest. The effect of imaging plane on the visualization of the left ventricle has not been fully described. To assess this effect, 10 normal adults underwent electrocardiogram-gated multiplanar cardiac MRI. The imaging planes being evaluated were standard planes, transverse, sagittal and coronal, and modified planes relating to the left ventricular (LV) long axis: coronal long axis parallel to the ventricular septum (VS), sagittal long axis perpendicular to the VS and transverse short axis perpendicular to the VS and long axis. The myocardium was separated into 9 segments. Using each plane, the LV segments were evaluated for quality of LV depiction. The planes based on the LV long axis were superior to standard planes in depicting the LV segments.

Phantom evaluation of imaging modalities for silicone breast implants.

RATIONALE AND OBJECTIVES: Recent concern regarding possible adverse effects from silicone breast implants has increased the role of radiologists in assessing augmented breasts. The authors compare the commonly available imaging modalities in evaluating the intact silicone implant as well as free silicone in the adjacent tissue. METHODS: A contrast resolution phantom and breast of veal phantom were tested. Fat was used as a reference material. The phantoms were imaged with xeromammography, film-screen mammography, ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI). Proton MRI spectroscopy also was performed on fat, silicone, water, and water/gelatin samples. The consensus of two radiologists determined whether free silicone was present. RESULTS: CT and MRI provided the best images of the implant and the free silicone. Several features of MRI were useful: spin-density scans and the fast low-angle shot (FLASH) and fast imaging with steady-state precision (FISP) techniques provided excellent resolution, a consistent chemical shift artifact appeared around the silicone, and frequency selective pre-saturation techniques resulted in marked suppression of the silicone. CONCLUSION: Additional testing in a more realistic setting, breast coil design, and improvement of various MRI techniques, particularly the frequency selective pre-saturation techniques, all appear promising in evaluating breast implants, the presence of free silicone, and the adjacent tissues.

Magnetic resonance imaging system stability: temporal variability in signal intensity, signal-to-noise, T1, and T2 measurements on a 0.15-T resistive system.

Signal intensity (SI) variability was evaluated on a 0.15-T resistive instrument using a phantom and was found to increase with time, repetition time (TR), and echo time (TE), ranging from 0.24% standard deviation (SD) over 34 min for 500/30 (TR/TE) images, to 2.1% SD over 5 days for 2000/30 images. Signal-to-noise (S/N) variability increased with time and TE but not TR and ranged from 4.2% SD over 34 min to 7.1% SD over 5 days in 500/30 images. Variability in T1 and T2 measurements on phantoms ranged from 1.8% to 4.8% SD for T1 and 3.6% to 6.5% SD for T2 in the biological range over 5 h. High reproducibility of SI, T1, and T2 measurements was demonstrated over a 6-week period in normal muscle measurements.

Slice profile improvement for a clinical MRI system.

Optimal control methods have been recently introduced to improve the design of selective radio frequency pulses and several optimized selective pulses that can produce excellent slice profiles have been reported. These pulses usually require high peak rf amplitudes to implement and thus can not be widely utilized because of the limitations of the specific absorption rate and the rf power amplifier of a clinical system. We have a Siemens 1.5 T MRI clinical system. Several pulse files which consist of the bandwidth matched 90 degrees and 180 degrees selective pulses are provided. Some of these can produce excellent slice profiles. However, they can only be used in the pulse sequences with the pulse length of 5.12 msec. The purpose of this paper is to improve the slice profiles produced by the pulse file in the pulse sequences with the shorter 2.56 msec pulse length. A pulse file optimized by the conjugate gradient method is proposed to substitute the 2.56 msec Siemens pulse file. Our experimental results confirm that the slice profiles and images are improved by the optimized pulse file with a lower peak voltage. The proposed pulse file can also be applied in other clinical MRI systems.

Barium sulfate suspension as a negative oral MRI contrast agent: in vitro and human optimization studies.

In vitro proton spectroscopy with line-width measurements and MR imaging were performed on various concentrations of commercially available single contrast (SC), double contrast, oral and rectal barium sulfate suspensions, as well as potassium sulfate, barium chloride, barium hydroxide, and 97% pure barium sulfate suspensions. Approximately 500 ml of 20%, 40%, 60%, and 70% w/w suspensions of SC oral barium sulfate suspensions were administered to four normal volunteers, respectively, and MR images were obtained at both 1.5 T and 0.15 T. Subsequently, 500 ml of 60% w/w suspensions of SC oral barium sulfate suspensions were administered to five normal volunteers and imaged at 1.5 T. All of the inert suspensions produced line-width broadening but the SC oral barium sulfate suspension at 50% and 70% stayed in suspension even after hours of standing undisturbed. As much as 80% of the small bowel and the entire colon were well visualized using the combination of 60% or 70% w/w SC barium sulfate suspensions with SE 550/22 and FISP pulse sequences. The effect was less at 0.15 T and also with the SE 2000/45/90 pulse sequences. We conclude that barium sulfate suspensions are useful as oral MRI contrast agents.

Magnetic resonance imaging of the pediatric thorax: initial experience.

Magnetic resonance imaging (MRI) is particularly useful in the pediatric patient. No ionizing radiation is used; there are no complicating side effects; the technique is noninvasive and painless. We reviewed our experience using a 0.15 T. resistive imager for diagnostic MRI of the pediatric thorax. We studied 16 children, aged 8 days to 16 years, with various thoracic diseases. Correlation of the MRI findings with computerized tomography (CT) was obtained in 10 children and with the surgical findings in 8. Spatial resolution was slightly less with MRI than with CT. Medium and large vessels are well seen without contrast medium injection. Tissue characterization with MRI is superior to CT and the normal thymus is well seen. The long scanning time can be a problem in young children but can be overcome using intramuscular sedation. Cardiac gated imaging was successfully tested.

Image quality and magnetic field strength in NMR imaging.

The rapid proliferation of nuclear magnetic resonance imagers necessitates an increasingly critical look at system requirements. Image quality in its broadest sense must be determined for various types of imaging systems. We review here the physical determinants of image quality and present examples of images acquired at several different field strengths.

Multisection-multiecho MR imaging: effect on image quality.

An analysis of the effects of multisection-multiecho imaging on signal intensity is presented using a gaussian excitation pulse. The number of sections acquired, size of the intersection gap, and the number of echoes were all varied independently to evaluate their effect on image intensity. The results indicate that one should take a conservative approach in multisection-multiecho imaging, leaving at least a 2-mm-intersection gap and approximately a 20-msec delay between section excitations to avoid image signal loss owing to partial saturation effects.

Double resonant quadrature birdcage.

The combined acquisition of proton images and localized spectra is considered essential to the application of NMR techniques to human and animal research. The ideal imaging/spectroscopy coil for our purposes would be one that provides the highest possible signal-to-noise, high homogeneity, and operation on two or more frequencies without retuning requirements or cable changes. To address these needs we have developed a quadrature double-tuned birdcage. We have incorporated our earlier work on the transformer coupled double-tuned surface coil into the birdcage structure by placing two birdcages in a coaxial configuration. This structure resonates at 34.6 MHz (phosphorus resonance at 2.0 T) and 85.5 MHz (proton resonance at 2.0 T). The quadrature performance of this coil for phosphorus was excellent, with a signal-to-noise that was 133% of our linear reference. The proton performance was less efficient, with a signal to noise that was 67% of our linear reference, but still quite sufficient for imaging. A phosphorus spectra and proton image of a rat abdomen are shown.

Varying gradient angles and offsets to optimize imaging planes in MR.

A simple technique for determining the appropriate viewing angle rotations necessary to obtain images in any desired plane is presented, along with an explanation of the imaging coordinates used in magnetic resonance imaging. Viewing angle rotations caused a minimal loss in image quality but no loss in signal-to-noise ratio, and were found to be compatible with both multisection and multiecho acquisitions. Several examples that show the practical applications of these techniques are provided.

Maximizing signal-to-noise ratio in the presence of coil coupling.

Crosstalk due to coupling produces noise correlation between receiver coils. It has been stated that this correlation reduces the signal-to-noise ratio obtainable from combining signals from the coils. In this paper, it is shown that the effects of crosstalk on the signal-to-noise ratio may in theory be eliminated by properly combining signals. Equations are derived which show how the signals from two coils should be combined in the presence of crosstalk in order to obtain the same signal-to-noise ratio as in an ideal case of no crosstalk. The deviation from optimum signal-to-noise ratio due to imperfect circuits and amplifiers is discussed. An experimental technique for achieving the proper combination of signals is presented.

Magnetic resonance imaging of fluid levels in an aneurysmal bone cyst and in anticoagulated human blood.

Magnetic resonance imaging (MRI) demonstrated a fluid level within an aneurysmal bone cyst (ABC). Since the ABC contained gross blood at operation, an anticoagulated human blood sample was studied by MRI also, and a fluid level was again clearly visible. MRI pulse sequences emphasizing T1 contrast showed the fluid levels most clearly in both the ABC and the blood. Sequences emphasizing T2 contrast showed homogeneous, bright signals in the ABC and in the blood, with no visible fluid level in the ABC and a nearly invisible one in the blood. In the blood sample, the calculated plasma T1 value was 1585 ms, and that of the red cells was 794 ms.

A transformer-coupled double-resonant probe for NMR imaging and spectroscopy.

A new type of double-resonant coil which takes advantage of the properties of a radio-frequency transformer is described. Two concentric loops are wound in close proximity to yield a high mutual inductance. The primary side is tuned to 34 MHz for 31P and the secondary side is tuned to 85 MHz for protons. A single lead allows program control over frequency without the need to rearrange cabling. Proton and phosphorus spectroscopy are made possible over essentially the same volume of interest without the need for repositioning the sample.

A comparison of double-tuned surface coils.

The combined acquisition of proton images and localized spectra is considered essential to the practical application of NMR techniques to human and animal research. Double-tuned surface coils which have been introduced to the literature are intended to address the problem; however, a careful evaluation of available designs is lacking. The "trap" method, the loop gap resonator design, and the transformer-coupled double-tuned design are evaluated here using bench tests of signal intensity and Q as well as signal-to-noise measurements on a 2-T imager/spectrometer. Comparisons are made relative to optimized single-tuned circuits of the same size for both protons at 85 MHz and phosphorus at 34 MHz. The results suggest that the "trap" design and the transformer coupled design are very efficient (98%) in the low-frequency mode (34 MHz) while the loop gap resonator is relatively inefficient (82%). In the high-frequency mode (85 MHz) the loop gap resonator is 75% efficient while the "trap" design and the transformer coupled coil are closer to 50% efficient.

A study of optimization of the complex hyperbolic secant inversion pulses.

The conjugate gradient method is used to optimize the inversion slice profile produced by complex hyperbolic secant selective pulses. An improved quality is obtained for the inversion slice profile at a price, however, of a lower immunity to the amplitude variation of radiofrequency pulses.

Proton imaging with surface coils on a 0.15-T resistive system.

Coil development is fundamental to the optimization of imaging systems at any field; however, it is especially critical for low-field systems where signal strength is limited. Here we demonstrate a type of surface coil design which makes possible the acquisition of high-quality images of the neck and extremities.

Advances in magnetic resonance imaging of the pelvis at 0.15 tesla.

The recent development of improved commercial radiofrequency coils and multiecho, multislice software for low field strength magnetic resonance systems has markedly increased the clinical utility of magnetic resonance imaging (MRI) of the pelvis at low field strengths. An evaluation of 70 patients with a variety of pelvic lesions and 14 normal volunteers who were studied using 0.15 T resistive magnet scanner revealed that anatomic structures and a variety of mass lesions could be clearly depicted in transaxial, sagittal and coronal planes using this updated system. Accurate characterization of lesions was possible in many instances using T2 weighted multiecho scans with echo time (TE) ranging from 30 ms to 120 ms (45 ms-180 ms using a reduced bandwidth technique). T1 weighted multislice scans demonstrated anatomic structures to best advantage and calculation of T1 and T2 relaxation times frequently facilitated more accurate differential diagnosis, particularly in the case of ovarian lesions.