MR systems for MRI-guided interventions.

The field of MR imaging has grown from diagnosis via morphologic imaging to more sophisticated diagnosis via both physiologic and morphologic imaging and finally to the guidance and control of interventions. A wide variety of interventional procedures from open brain surgeries to noninvasive focused ultrasound ablations have been guided with MR and the differences between diagnostic and interventional MR imaging systems have motivated the creation of a new field within MR. This review discusses the various systems that research groups and vendors have designed to meet the requirements of interventional MR and suggest possible solutions to those requirements that have not yet been met. The common requirements created by MR imaging guidance of interventional procedures are reviewed and different imaging system designs will be independently considered. The motivation and history of the different designs are discussed and the ability of the designs to satisfy the requirements is analyzed.

Human meniscus allografts' in vivo size and motion characteristics: magnetic resonance imaging assessment under weightbearing conditions.

BACKGROUND: Although many reports have described clinical outcome of meniscus transplants, their size and motion patterns are not well understood. This study assessed postoperative in vivo size and motion patterns of meniscus transplants under full weightbearing conditions. HYPOTHESIS: The human meniscus transplant has size and motion characteristics similar to the native menisci. STUDY DESIGN: Case series; Level of evidence, 4. METHODS: Eight meniscus transplants in 7 knees were studied 15 to 34 months postoperatively. The knees were scanned in an open magnetic resonance imaging 0.5-T superconducting magnet while standing. Single slice sagittal and coronal images were obtained at 0 degrees , 30 degrees , 60 degrees , and 90 degrees of flexion in the operative and contralateral knees. RESULTS: The mean height and width of the anterior and posterior horns of the transplants were similar to those of native menisci. The millimeters of coronal displacement of motion of the midbody were also similar between the transplants and the native menisci. The anterior horn of the native medial menisci moved a mean of 5 mm more (total anterior to posterior translation, P < .05) than did the transplants. The posterior horn of the native medial menisci, and both horns of the native lateral menisci, also tended to move more than the corresponding horns of the transplanted menisci, although this result could not be confirmed statistically, given the number of menisci studied. CONCLUSION: Meniscus transplants had similar size and midbody motion characteristics as the native menisci. The horns of the meniscus transplants tended to show decreased motion compared with the native menisci. The operative techniques and subsequent healing of the graft bone and peripheral attachments provided a stable meniscus construct.

Quantifying the effect of posture on intracranial physiology in humans by MRI flow studies.

PURPOSE: To quantify the effect of posture on intracranial physiology in humans by MRI, and demonstrate the relationship between intracranial compliance (ICC) and pressure (ICP), and the pulsatility of blood and CSF flows. MATERIALS AND METHODS: Ten healthy volunteers (29+/-7 years old) were scanned in the supine and sitting positions using a vertical gap MRI scanner. Pulsatile blood and CSF flows into and out from the brain were visualized and quantified using time-of-flight (TOF) and cine phase-contrast techniques, respectively. The total cerebral blood flow (tCBF), venous outflow, ICC, and ICP for the two postures were then calculated from the arterial, venous, and CSF volumetric flow rate waveforms using a previously described method. RESULTS: In the upright posture, venous outflow is considerably less pulsatile (57%) and occurs predominantly through the vertebral plexus, while in the supine posture venous outflow occurs predominantly through the internal jugular veins. A slightly lower tCBF (12%), a considerably smaller CSF volume oscillating between the cranium and the spinal canal (48%), and a much larger ICC (2.8-fold) with a corresponding decrease in the MRI-derived ICP values were measured in the sitting position. CONCLUSION: The effect of posture on intracranial physiology can be quantified by MRI because posture-related changes in ICC and ICP strongly affect the dynamics of cerebral blood and CSF flows. This study provides important insight into the coupling that exists between arterial, venous, and CSF flow dynamics, and how it is affected by posture.

Safety protocols for interventional MRI.

RATIONALE AND OBJECTIVES: To review magnetic resonance safety protocols and supplement them for interventional applications. MATERIALS AND METHODS: The American College of Radiology White Papers on MR Safety are summarized. Elements relevant to interventional MR procedures are emphasized. Additional policies and procedures specifically for interventional MR applications covering safety, training, and MR compatibility are proposed. RESULTS: A comprehensive operational policy for an interventional MR department is proposed. The policies and procedures helped our department perform over 400 surgeries and maintain a perfect safety record for 5 years. CONCLUSION: Interventional MR departments can operate safely but require policies and procedures beyond those required for diagnostic MR departments.

Automatic cerebrovascular segmentation by accurate probabilistic modeling of TOF-MRA images.

Accurate automatic extraction of a 3D cerebrovascular system from images obtained by time-of-flight (TOF) or phase contrast (PC) magnetic resonance angiography (MRA) is a challenging segmentation problem due to small size objects of interest (blood vessels) in each 2D MRA slice and complex surrounding anatomical structures, e.g. fat, bones, or grey and white brain matter. We show that due to a multi-modal nature of MRA data blood vessels can be accurately separated from background in each slice by a voxel-wise classification based on precisely identified probability models of voxel intensities. To identify the models, an empirical marginal probability distribution of intensities is closely approximated with a linear combination of discrete Gaussians (LCDG) with alternate signs, and we modify the conventional Expectation-Maximization (EM) algorithm to deal with the LCDG. To validate the accuracy of our algorithm, a special 3D geometrical phantom motivated by statistical analysis of the MRA-TOF data is designed. Experiments with both the phantom and 50 real data sets confirm high accuracy of the proposed approach.

Dynamic weight-bearing cervical magnetic resonance imaging: technical review and preliminary results.

BACKGROUND: Conventional magnetic resonance imaging (MRI) of complex cervical spine disorders may underestimate the magnitude of structural disease because imaging is performed in a nondynamic non-weight-bearing manner. Myelography provides additional information but requires an invasive procedure. METHODS: This was a prospective review of the first 20 upright weight-bearing cervical MRI procedures with patients in the flexed, neutral, and extended positions conducted in an open-configuration MRI unit. RESULTS: This technique clearly illustrated the changes in spinal cord compression, angulation, and spinal column alignment that occur during physiologic movements with corresponding changes in midsagittal spinal canal diameter (P < 0.05). Image quality was excellent or good in 90% of the cases. CONCLUSIONS: Dynamic weight-bearing MRI provides an innovative method for imaging complex cervical spine disorders. This technique is noninvasive and has adequate image quality that may make it a good alternative to cervical myelography.

3D volume segmentation of MRA data sets using level sets: image processing and display.

In this article, we use a level set-based segmentation algorithm to extract the vascular tree from magnetic resonance angiography (MRA) data sets. The classification approach depends on initializing the level sets in the 3D volume, and the level sets evolve with time to yield the blood vessels. This work introduces a high-quality initialization for the level set functions, allowing extraction of the blood vessels in 3D and elimination of non-vessel tissues. A comparison between the 2D and 3D segmentation approaches is made. The results are validated using a phantom that simulates the MRA data and show good accuracy.

Magnetic resonance imaging analysis of the subacromial space in the impingement sign positions.

External shoulder impingement is commonly diagnosed by passively moving the shoulder into various positions of elevation and internal rotation in order to perform impingement sign maneuvers. There is a lack of agreement among clinicians regarding the positions of the anatomic structures in the subacromial space when these maneuvers are performed. The purpose of this study was to use magnetic resonance imaging to identify and measure the changes in anatomic structures in the subacromial space as the arm was moved from complete rest to 160 degrees of forward flexion during the Neer and Hawkins impingement sign maneuvers. Ten subjects with normal shoulders (mean age, 32 years) were studied. The acromiohumeral interval was smallest with the arm at the side (mean, 6.4 mm) and progressively increased as the arm was elevated from 90 degrees to 160 degrees (mean, 7.7-14.2 mm). In no instance was the rotator cuff found to be in contact with the anterior acromion. The rotator cuff insertion appeared to be in closest proximity to the anteroinferior acromion, not at full elevation (Neer sign position), but at 90 degrees of flexion (Hawkins sign position). Our data suggest that a clinically positive Hawkins sign is consistent with external shoulder impingement. Clinical suspicion that mechanisms other than impingement may be involved, particularly if the Neer impingement sign is also positive, is nonetheless advised.

Interventional MRI-guided frameless stereotaxy in pediatric patients.

INTRODUCTION: We prospectively reviewed our experience with intraoperative MRI (iMRI)-guided stereotactic procedures in pediatric patients. METHODS: All procedures were performed within the magnet bore of the General Electric Signa SP MRI system, which allows for either continuous real-time or periodic imaging. The internal optical tracking system was used to plan and monitor target localization and instrument trajectory. RESULTS: Fifteen patients underwent 16 frameless stereotactic procedures, consisting of 4 tumor biopsies and 12 cyst aspirations and stereotactic catheter placements (average age 6 years, range 6 weeks to 18 years). There were no hemorrhagic, neurologic or infectious complications. CONCLUSION: iMRI is an important component in expanding the horizon of minimally invasive neurosurgery for pediatric patients. Thus far, we have found this technology to be safe, reliable and extremely useful for frameless stereotactic procedures.