MRI (magnetic resonance imaging) with ¹²⁹Xe has gained much attention as a diagnostic methodology because of its affinity for lipids and possible polarization. The quantitative estimation of net detectability and stability of hyperpolarized ¹²⁹Xe in the dissolved phase in vivo is valuable to the development of clinical applications. The goal of this study was to develop a stable hyperpolarized ¹²⁹Xe experimental 3T system to statistically analyze the dissolved-phase ¹²⁹Xe signal in the rat lungs.
The polarization of ¹²⁹Xe with buffer gases at the optical pumping cell was measured under adiabatic fast passage against the temperature of an oven and laser absorption at the cell. The gases were insufflated into the lungs of Sprague-Dawley rats (n=15, 400-550 g) through an endotracheal tube under spontaneous respiration. Frequency-selective spectroscopy was performed for the gas phase and dissolved phase. We analyzed the ¹²⁹Xe signal in the dissolved phase to measure the chemical shift, T₂^*, delay and its ratio in a rat lungs on 3T.
The polarizer was able to produce polarized gas (1.1±0.47%, 120 cm³) hundreds of times with the laser absorption ratio (25%) kept constant at the cell. The optimal buffer gas ratio of 25-50% rendered the maximum signal in the dissolved phase. Two dominant peaks of 211.8±0.9 and 201.1±0.6 ppm were observed with a delay of 0.4±0.9 and 0.9±1.0 s from the gas phase spectra. The ratios of their average signal to that of the gas phase were 5.6±5.2% and 4.4±4.7%, respectively. The T₂^* of the air space in the lungs was 2.5±0.5 ms, which was 3.8 times shorter than that in a syringe.
We developed a hyperpolarized ¹²⁹Xe experimental system using a 3T MRI scanner that yields sufficient volume and polarization and quantitatively analyzed the dissolved-phase ¹²⁹Xe signal in the rat lungs.

View full abstract

Purpose: In order to ensure that three-dimensional diffusion tensor tractography (3D-DTT) of the corticospinal tract (CST), is performed accurately and efficiently, we set out to find the optimal lower threshold of fractional anisotropy (FA) below which tract elongation is terminated (trackability threshold).
Methods: Thirteen patients with acute or early subacute ischemic stroke causing motor deficits were enrolled in this study. We performed 3D-DTT of the CST with diffusion tensor MR (magnetic resonance) imaging. We segmented the CST and established a cross-section of the CST in a transaxial plane as a region of interest. Thus, we selectively measured the FA values of the right and left corticospinal tracts at the level of the cerebral peduncle, the posterior limb of the internal capsule, and the centrum semiovale. The FA values of the CST were also measured on the affected side at the level where the clinically relevant infarction was present in isotropic diffusion-weighted imaging.
Results: 3D-DTT allowed us to selectively measure the FA values of the CST. Among the 267 regions of interest we measured, the minimum FA value was 0.22. The FA values of the CST were smaller and more variable in the centrum semiovale than in the other regions. The mean minus twice the standard deviation of the FA values of the CST in the centrum semiovale was calculated at 0.22 on the normal unaffected side and 0.16 on the affected side.
Conclusion: An FA value of about 0.20 was found to be the optimal trackability threshold.

View full abstract

Creatine depletion in the non-viable infarcted human heart was previously demonstrated with proton magnetic resonance (MR) spectroscopy (¹H MRS). In the present study, we assessed total creatine (CR) in human hearts with non-ischemic dysfunctions such as cardiomyopathy. Using cardiac-gated ¹H MRS with MR image-guided PRESS localization, we measured septal CR in healthy and diseased human hearts. Fifteen patients with chronic heart failure (CHF, left ventricular ejection fraction <45%) and 14 age-matched normal subjects were examined. Myocardial CR was significantly (p<0.001) lower in failing hearts (15.1±SD 5.0 μmol/g wet weight, range 8.0-22.9) than in normal hearts (27.6±4.1 μmol/g wet weight, range 20.8-36.2). Myocardial CR concentrations in six heart failure patients with plasma B-type natriuretic peptide (BNP) levels of >200 pg/ml (11.5±0.9 μmol/g wet weight, range 9.9-12.3) were significantly lower than those in four heart failure patients with plasma BNP levels of <200 pg/ml (19.8±2.5 μmol/g wet weight, range 17.7-22.9, p<0.001). Thus, our study showed that myocardial CR was decreased in non-ischemic dysfunctional hearts. Noninvasive measurements of myocardial CR by ¹H MRS may be useful in the assessment of the severity of heart failure.

View full abstract

The evident advantage of high-field MR (magnetic resonance) scanners is their higher signal-to-noise ratio, which results in improved imaging. While no reliable efficacy studies exist that compare the diagnostic capabilities of low- versus high-field scanners, the adoption and acceptance of low-field MRI (magnetic resonance imaging) is subject to biases. On the other hand, the cost savings associated with low-field MRI hardware are obvious. The running costs of a non-superconductive low-field scanner show even greater differences in favor of low-field scanners. Patient anxiety and safety issues also reflect the advantages of low-field scanners. Recent technological developments in the realm of low-field MR scanners will lead to higher image quality, shorter scan times, and refined imaging protocols. Interventional and intraoperative use also supports the installation of low-field MR scanners. Utilization of low-field systems has the potential to enhance overall cost reductions with little or no loss of diagnostic performance.

View full abstract

Purpose: To compare the performance of the phased-array coil (PAC) with that of the single-loop coil (SLC) in magnetic resonance (MR) imaging of the breast.
Materials and Methods: MR imaging was performed with a 1.5T MR imager. A phantom study was performed with the right element of the two coils to obtain their signal-to-noise ratio (SNR). MR images of the breasts of 12 patients with breast lesions were obtained with the PAC and SLC, and these images were reviewed by five readers in a blind evaluation employing a scoring system for assessing overall image quality.
Results: In the phantom study, the SLC exhibited a SNR 1.82 times higher than that of the PAC at the center of the coil; however, the SLC exhibited an inhomogeneous sensitivity profile and its SNR varied with the distance from the center of the coil in the horizontal and vertical directions. In most of the 12 patients, the MR images obtained with the PAC showed more noise than did those obtained with the SLC, and the PAC obtained lower scores than the SLC in the assessment of overall image quality; however, the difference was significant (p<0.05) only in coronal imaging. On the other hand, the uniformity of fat saturation in the MR images obtained with the PAC was judged to be significantly superior to that obtained with the SLC (p<0.05).
Conclusion: Compared with the SLC, the PAC exhibited a lower SNR and was less advantageous at depicting the breast. However, the PAC provided more homogeneous fat saturation and might be useful for reducing artifacts.

View full abstract

Background: Parallel imaging can be applied to cardiac imaging with a cylindrical MRI (magnetic resonance imaging) apparatus. Studies of open MRI, however, are few. This study sought to achieve cardiac cine parallel imaging (or RAPID, for “rapid acquisition through parallel imaging design”) with an open 0.7T MRI apparatus.
Materials and Methods: Imaging time was shortened in all slice directions with the use of a dedicated four-channel RF receiving coil comprising solenoid coils and butterfly coils. Coil shape was designed through an RF-coil simulation that considered biological load. The auto-calibration of a 0.7T open MRI apparatus incorporated a modified image-domain reconstruction algorithm. Cine images were obtained with a BASG, or balanced SARGE (steady-state acquisition with rewound gradient echo), sequence. Image quality was evaluated with cylindrical phantoms and five healthy volunteers.
Results: Multi-slice phantom images showed no visible artifacts. Cine images taken under breath-hold with an acceleration factor of two were evaluated carefully. With auto-calibration, the images revealed no visible unfolded artifacts or motion artifacts. RAPID thus improved the acquisition speed, time resolution, and spatial resolution of short-axis, long-axis, and four-chamber images.
Conclusion: The use of a dedicated RF coil enabled cardiac cine RAPID to be performed with an open MRI apparatus.

View full abstract