The utility of pelvic coil SNR testing in the quality assurance of a clinical MRgFUS system.

During MRI-guided focused ultrasound (MRgFUS) treatments of uterine fibroids using ExAblate 2000, tissue ablations are delivered by a FUS transducer while MR imaging is performed with a pelvic receiver coil. The consistency of the pelvic coil performance is crucial for reliable MR temperature measurements as well as detailed anatomic imaging in patients. Test sonications in a gel phantom combined with MR thermometry are used to test the performance of the FUS transducer prior to each treatment. As we show, however, these tests do not adequately evaluate receiver coil performance prior to clinical use. This could become a problem since the posterior part of the coil is frequently moved and can malfunction. The aim of this work is to demonstrate the utility of the signal-to-noise ratio (SNR) as a reliable indicator of pelvic coil performance. Slight modification of the vendor-provided coil support was accomplished to assure reproducible coil positioning. The SNR was measured in a gel phantom using axial acquisitions from the 3D-localizer scan. MR temperature and SNR measurements were obtained using a degraded receiver coil (with posterior element removed) and a known faulty coil, and compared to those obtained with a fully functioning coil. While the MR temperature-based tests were insensitive to change in pelvic coil performance, (degraded, p = 0.24; faulty, p = 0.28), the SNR tests were highly sensitive to coil performance, (degraded, p < 0.001; faulty, p < 0.001). Additional clinical data illustrate the utility of SNR testing of the receiver coil. These tests require minimal (or possibly no) additional scan time and have proven to be effective in our clinical practice.

Calorimetric calibration of head coil SAR estimates displayed on a clinical MR scanner.

Calorimetric measurements were performed to determine the average specific absorption rates (SAR) resulting from MRI head examinations. The data were compared with average head coil SAR estimates displayed by the MR scanner in order to refine the imaging protocols used in imaging patients with implanted deep brain stimulators (DBS). The experiments were performed using transmit-receive (TR) head coil on clinical 1.5 T General Electric MR scanners running 11.0 M4 revision software. The average applied SAR was derived from temperature increases measured inside a head phantom, due to deposition of RF energy during MRI scanning with a spin echo imaging sequence. The measurements were repeated for varied levels of RF transmit gain (TG) and analyzed with a range of entered patient weights. The measurements demonstrate that the ratio of the actual average head SAR to the scanner-displayed value (coil correction factor) decreases for decreasing TG or for increasing patient weight and may vary between 0.3 and 2.1. An additional retrospective patient study, however, shows that not all combinations of TG and patient weight are encountered clinically and, instead, TG generally increases with the patient weight. As a result, a much narrower range of coil correction factors (e.g., typically 0.5-1.0) will be encountered in practice. The calorimetric method described in this work could aid the physicians and technologists in refinement of the model-dependent SAR estimates displayed by the MR scanner, and in selection of imaging parameters for MR head examinations within allowable SAR safety levels.

The Alzheimer's Disease Neuroimaging Initiative (ADNI): MRI methods.

The Alzheimer's Disease Neuroimaging Initiative (ADNI) is a longitudinal multisite observational study of healthy elders, mild cognitive impairment (MCI), and Alzheimer's disease. Magnetic resonance imaging (MRI), (18F)-fluorodeoxyglucose positron emission tomography (FDG PET), urine serum, and cerebrospinal fluid (CSF) biomarkers, as well as clinical/psychometric assessments are acquired at multiple time points. All data will be cross-linked and made available to the general scientific community. The purpose of this report is to describe the MRI methods employed in ADNI. The ADNI MRI core established specifications that guided protocol development. A major effort was devoted to evaluating 3D T(1)-weighted sequences for morphometric analyses. Several options for this sequence were optimized for the relevant manufacturer platforms and then compared in a reduced-scale clinical trial. The protocol selected for the ADNI study includes: back-to-back 3D magnetization prepared rapid gradient echo (MP-RAGE) scans; B(1)-calibration scans when applicable; and an axial proton density-T(2) dual contrast (i.e., echo) fast spin echo/turbo spin echo (FSE/TSE) for pathology detection. ADNI MRI methods seek to maximize scientific utility while minimizing the burden placed on participants. The approach taken in ADNI to standardization across sites and platforms of the MRI protocol, postacquisition corrections, and phantom-based monitoring of all scanners could be used as a model for other multisite trials.

Nonlinear averaging reconstruction method for phase-cycle SSFP.

The ability to obtain high-quality images of small structures, such as the nerves of the inner ear, is important for the early diagnosis of numerous conditions. Balanced steady-state free precession (SSFP; e.g., true fast imaging with steady-state precession) is a fast acquisition method, but its use has been limited by the presence of off-resonance banding artifacts. To reduce these artifacts multiacquisition balanced SSFP with phase cycling is used, yielding multiple data sets in which the banding artifacts are spatially shifted with respect to each other (e.g., as in CISS). We present a new method, called nonlinear averaging (NLA), for combining these data sets to reduce banding artifacts. The NLA method arithmetically averages the three highest magnitude signals from four-phase-cycle SSFP data on a pixel-by-pixel basis. Simulations indicate that NLA offers improved signal-to-noise ratio (SNR) over the more standard maximum intensity projection (MIP) reconstruction. NLA is compared to MIP in simulations and volunteer tests. Simulations suggest that NLA provides substantially improved SNR compared to MIP. In a randomized blinded comparison of 10 volunteer studies, two radiologists found that NLA, compared to MIP, gave improved results. NLA also provided superior noise reduction and enhanced edge sharpness compared to MIP. We demonstrate that NLA, similar to MIP, improves SNR and image quality. It does so consistently in all situations to which it is applied.

Imaging artifacts at 3.0T.

Clinical MRI at a field strength of 3.0T is finding increasing use. However, along with the advantages of 3.0T, such as increased SNR, there can be drawbacks, including increased levels of imaging artifacts. Although every imaging artifact observed at 3.0T can also be present at 1.5T, the intensity level is often higher at 3.0T and thus the artifact is more objectionable. This review describes some of the imaging artifacts that are commonly observed with 3.0T imaging, and their root causes. When possible, countermeasures that reduce the artifact level are described.

Rapid T2 estimation with phase-cycled variable nutation steady-state free precession.

Variable nutation SSFP (DESPOT2) permits rapid, high-resolution determination of the transverse (T2) relaxation constant. A limitation of DESPOT2, however, is the presence of T2 voids due to off-resonance banding artifacts associated with SSFP images. These artifacts typically occur in images acquired with long repetition times (TR) in the presence of B0 inhomogeneities, or near areas of magnetic susceptibility difference, such that the transverse magnetization experiences a net phase shift during the TR interval. This places constraints on the maximum spatial resolution that can be achieved without artifact. Here, a novel implementation of DESPOT2 is presented incorporating RF phase-cycling which acts to shift the spatial location of the bands, allowing reconstruction of a single, reduced artifact-image. The method is demonstrated in vivo with the acquisition of a 0.34 mm3 isotropic resolution T2 map of the brain with high precision and accuracy and significantly reduced artifact.

3-T imaging of the cochlear nerve and labyrinth in cochlear-implant candidates: 3D fast recovery fast spin-echo versus 3D constructive interference in the steady state techniques.

BACKGROUND AND PURPOSE: High-resolution imaging of the internal auditory canal and labyrinth at 1.5 T is often performed by using three-dimensional (3D) fast spin-echo or T2* techniques. We evaluated both techniques at 3 T in the preoperative assessment of patients being considered for cochlear implants. METHODS: Sagittal 3D fast recovery fast spin-echo (FRFSE) and 3D constructive interference in the steady state (CISS) images were acquired in eight patients at 3.0 T by using dual surface coils. Contrast-to-noise ratios (CNRs) for the intracanalicular nerve and CSF were measured in the internal auditory canal. Two neuroradiologists reviewed the images to determine whether the techniques provided images of diagnostic quality. RESULTS: CNRs for 3D CISS were twice those obtained with 3D FRFSE. Both techniques provided images of diagnostic quality, though spurious signal intensity loss at the apex of the superior semicircular canals was encountered on 3D FRFSE images in four of eight patients. CONCLUSION: Both 3D FRFSE and 3D CISS provide high-resolution images of the internal auditory canal and labyrinth at 3.0 T. We predict that the superior CNRs obtained with 3D CISS will prove advantageous as we move to smaller fields of view at higher field strength.

Real-time autoshimming for echo planar timecourse imaging.

Head motion within an applied magnetic field alters the effective shim within the brain, causing geometric distortions in echo planar imaging (EPI). Even if subtle, change in shim can lead to artifactual signal changes in timecourse EPI acquisitions, which are typically performed for functional MRI (fMRI) or diffusion tensor imaging. Magnetic field maps acquired before and after head motions of clinically realistic magnitude indicate that motion-induced changes in magnetic field may cause translations exceeding 3 mm in the phase-encoding direction of the EPI images. The field maps also demonstrate a trend toward linear variations in shim changes as a function of position within the head, suggesting that a real-time, first-order correction may compensate for motion-induced changes in magnetic field. This article presents a navigator pulse sequence and processing method, termed a "shim NAV," for real-time detection of linear shim changes, and a shim-compensated EPI pulse sequence for dynamic correction of linear shim changes. In vivo and phantom experiments demonstrate the detection accuracy of shim NAVs in the presence of applied gradient shims. Phantom experiments demonstrate reduction of geometric distortion and image artifact using shim-compensated EPI in the presence of applied gradient shims. In vivo experiments with intentional interimage subject motion demonstrate improved alignment of timecourse EPI images when using the shim NAV-detected values to update the shim-compensated EPI acquisition in real time.

Spherical navigator echoes for full 3D rigid body motion measurement in MRI.

We developed a 3D spherical navigator (SNAV) echo technique that can measure rigid body motion in all six degrees of freedom simultaneously by sampling a spherical shell in k-space. 3D rotations of an imaged object simply rotate the data on this shell and can be detected by registration of k-space magnitude values. 3D translations add phase shifts to the data on the shell and can be detected with a weighted least-squares fit to the phase differences at corresponding points. MRI pulse sequences were developed to study k-space sampling strategies on such a shell. Data collected with a computer-controlled motion phantom with known rotational and translational motions were used to evaluate the technique. The accuracy and precision of the technique depend on the sampling density. Roughly 2000 sample points were necessary for accurate detection to within the error limits of the motion phantom when using a prototype time-intensive sampling method. This number of samples can be captured in an approximately 27-ms double excitation SNAV pulse sequence with a 3D helical spiral trajectory. Preliminary results with the helical SNAV are encouraging and indicate that accurate motion measurement suitable for retrospective or prospective correction should be feasible with SNAV echoes.