Reiner Umathum

MR-guided intravascular procedures : real-time parameter control and automated slice positioning with active tracking coils

To implement and optimize a real‐time pulse sequence and user interface to perform intravascular interventions using active catheter tracking.

MR coil design for simultaneous tip tracking and curvature delineation of a catheter

In active catheter tracking, small RF coils are attached to the catheter for localization. For interactive catheter steering at vessel branchings, it is necessary to visualize not only a single point near the catheter tip but also the entire shape and orientation of the catheters distal end. Therefore, a 35‐mm‐long twisted‐pair RF coil was added to a 5 French intravascular catheter with a single tip‐tracking coil. With the use of small nonmagnetic electronic components at the catheter tip, and a special switching circuitry outside the catheter, the coil assembly could be operated in two different modes. During MRI, the tip‐tracking coil was detuned so that the MR signal was received by the visualization coil only. During tracking, detuning was switched off and the MR signal was predominantly received by the more sensitive tracking coil. The catheter was used in combination with a MR pulse sequence with automatic slice positioning so that the current imaging slice was always placed at the position of the catheter tip. Phantom and animal experiments showed that the catheter tip is better visualized with the combined approach than with a tracking coil alone. Magn Reson Med 52:214–218, 2004.

Active catheter tracking using parallel MRI and real-time image reconstruction.

In this work active MR catheter tracking with automatic slice alignment was combined with an autocalibrated parallel imaging technique. Using an optimized generalized autocalibrating partially parallel acquisitions (GRAPPA) algorithm with an acceleration factor of 2, we were able to reduce the acquisition time per image by 34%. To accelerate real‐time GRAPPA image reconstruction, the coil sensitivities were updated only after slice reorientation. For a 2D trueFISP acquisition (160 × 256 matrix, 80% phase matrix, half Fourier acquisition, TR = 3.7 ms, GRAPPA factor = 2) real‐time image reconstruction was achieved with up to six imaging coils. In a single animal experiment the method was used to steer a catheter from the vena cava through the beating heart into the pulmonary vasculature at an image update rate of about five images per second. Under all slice orientations, parallel image reconstruction was accomplished with only minor image artifacts, and the increased temporal resolution provided a sharp delineation of intracardial structures, such as the papillary muscle. Magn Reson Med, 2006.

Manganese-enhanced magnetic resonance imaging for in vivo assessment of damage and functional improvement following spinal cord injury in mice

In past decades, much effort has been invested in developing therapies for spinal injuries. Lack of standardization of clinical read‐out measures, however, makes direct comparison of experimental therapies difficult. Damage and therapeutic effects in vivo are routinely evaluated using rather subjective behavioral tests. Here we show that manganese‐enhanced magnetic resonance imaging (MEMRI) can be used to examine the extent of damage following spinal cord injury (SCI) in mice in vivo. Injection of MnCl2 solution into the cerebrospinal fluid leads to manganese uptake into the spinal cord. Furthermore, after injury MEMRI‐derived quantitative measures correlate closely with clinical locomotor scores. Improved locomotion due to treating the detrimental effects of SCI with an established therapy (neutralization of CD95Ligand) is reflected in an increase of manganese uptake into the injured spinal cord. Therefore, we demonstrate that MEMRI is a sensitive and objective tool for in vivo visualization and quantification of damage and functional improvement after SCI. Thus, MEMRI can serve as a reproducible surrogate measure of the clinical status of the spinal cord in mice, potentially becoming a standard approach for evaluating experimental therapies. Magn Reson Med, 2006.

Renal embolization: feasibility of magnetic resonance-guidance using active catheter tracking and intraarterial magnetic resonance angiography.

Rationale and Objectives:Magnetic resonance (MR)-guidance of endovascular interventions offers various advantages, including the absence of ionizing radiation, excellent soft tissue contrast, and multiplanar and functional imaging capabilities. The objective of this study was to assess the feasibility of MR-guided renal embolization using active catheter tracking with automatic slice positioning and intraarterial contrast-enhanced MR angiography (MRA). Materials and Methods:MR-guided embolization of 16 kidneys was attempted in 15 pigs using real-time tracking of active 5-Fr. catheters. Embolization was monitored by selective intraarterial projection MRA. Intraarterial three-dimensional (3D) MRA was used for the assessment of embolization results. Additional pathologic correlation was available in 2 animals. The image quality of intraarterial 3D contrast-enhanced-MRA was rated by an independent radiologist who was not involved in the animal experiments. Results:Active catheter tracking with automatic slice positioning allowed reliable catheter guidance and catheterization of the renal artery in all animals. Embolization was successful in all kidneys (11 left, 5 right), as verified by intraarterial 3D contrast-enhanced MRA (ce-MRA) and/or pathology. The image quality of intraarterial 3D ce-MRA was rated excellent in 10 animals, moderate in 4 animals, and poor in 1 animal. Conclusion:Renal embolization using active catheter tracking and intraarterial ce-MRA is feasible. Selective intraarterial ce-MRA allows the assessment of blood supply and organ perfusion before, during, and after therapeutic interventions, thereby complementing MR-guided endovascular interventions.

A measurement setup for direct 17O MRI at 7 T

An efficient breathing system was designed for direct 17O MRI to perform oxygen metabolism studies of the human brain. The breathing system consists of a demand oxygen delivery device for 17O2 supply and a custom‐built re‐breathing circuit with pneumatic switching valve. To efficiently deliver the 17O gas to the alveoli of the lungs, the system applies short gas pulses upon an inspiration trigger via a nasal cannula. During and after 17O2 administration, the exhaled gas volumes are stored and filtered in the re‐breathing section to make the most efficient use of the rare 17O gas. In an inhalation experiment, 2.2 ± 0.1 L of 70%‐enriched 17O2 were administered to a healthy volunteer and direct 17O MRI was performed for a total imaging time of 38 min with a temporal resolution of 50 s per 3D data set. Mapping of the maximum signal increase was carried out showing regional variations of oxygen concentration of up to 30% over the natural abundance of 17O water. Magn Reson Med, 2011.

Targeted-HASTE imaging with automated device tracking for MR-guided needle interventions in closed-bore MR systems

Percutaneous MR‐guided interventions with needles require fast pulse sequences to image the needle trajectory with minimal susceptibility artifacts. Spin‐echo pulse sequences are well suited for reducing artifact size; however, even with single‐shot turbo spin‐echo techniques, such as rapid acquisition with relaxation enhancement (RARE) or half‐Fourier acquisition single‐shot turbo spin‐echo (HASTE), fast imaging remains challenging. In this work we present a HASTE pulse sequence that is combined with inner‐volume excitation to reduce the scan time and limit the imaging field of view (FOV) to a small strip close to the needle trajectory (targeted‐HASTE). To compensate for signal saturation from fast repeated acquisitions, a magnetization restore pulse (driven equilibrium Fourier transform (DEFT)) is used. The sequence is combined with dedicated active marker coils to measure the position and orientation of the needle so that the targeted‐HASTE image slice is automatically repositioned. In an animal experiment the coils were attached to an MR‐compatible robotic assistance system for MR‐guided interventions. Needle insertion and infusion via the needle could be visualized with a temporal resolution of 1 s, and the needle tip could be localized even in the presence of a stainless steel mandrel. Magn Reson Med, 2006.

In Vivo 35Cl MR Imaging in Humans: A Feasibility Study

PURPOSEnTo implement chlorine 35 ((35)Cl) magnetic resonance (MR) at a 7-T whole-body MR system and evaluate its feasibility for imaging humans.nnnMATERIALS AND METHODSnAll examinations were performed with ethical review board approval; written informed consent was obtained from all volunteers. Seven examinations each of brain and muscle in healthy volunteers and four examinations of patients were performed. Two patients with histologically confirmed glioblastoma multiforme underwent brain imaging. (35)Cl MR and (35)Cl inversion-recovery (IR) MR were performed. Two patients with genetically confirmed hypokalemic periodic paralysis underwent calf muscle imaging. Seven multiecho sequences (acquisition time, 5 minutes; voxel dimension, 11 mm(3)) were applied to determine transverse relaxation time as affected by magnetic field heterogeneity (T2*) and chlorine concentration. (35)Cl and sodium 23 ((23)Na) MR were conducted with a 7-T whole-body MR system. (35)Cl longitudinal relaxation time (T1) and T2* of healthy human brain and muscle were determined with a three-dimensional density-adapted-projection reconstruction technique to achieve short echo times and high signal-to-noise ratio (SNR) efficiency. A nonlinear least squares routine and mono- (T1) and biexponential (T2*) models were used for curve fitting.nnnRESULTSnPhantom imaging revealed 15-fold lower SNR and much shorter relaxation times for (35)Cl than (23)Na. In vivo T2* was biexponential and extremely short. Monoexponential fits of T1 revealed 9.2 and 4.0 milliseconds ± 0.7 (standard deviation) for brain and muscle, respectively. In glioblastoma tissue, increased Cl(-) concentrations and increased Cl(-) IR signal intensities were detected. Voxel dimension and acquisition time, respectively, were 6 mm(3) and 9 minutes 45 seconds ((35)Cl MR) and 10 mm(3) and 10 minutes ((35)Cl IR MR). In patients with hypokalemic periodic paralysis versus healthy volunteers, Cl(-) and Na(+) concentrations were increased. Cl(-) concentration of muscle could be determined (voxel size, 11 mm(3); total acquisition time, 35 minutes).nnnCONCLUSIONnMR at 7 T enables in vivo imaging of (35)Cl in human brain and muscle in clinically feasible acquisition times (10-35 minutes) and voxel volumes (0.2-1.3 cm(3)). Pathophysiological changes of Cl(-) homeostasis due to cancer or muscular ion channel disease can be visualized.

B1 field-insensitive transformers for RF-safe transmission lines

Objective: Integration of transformers into transmission lines suppresses radiofrequency (RF)-induced heating. New figure-of-eight-shaped transformer coils are compared to conventional loop transformer coils to assess their signal transmission properties and safety profile.Materials and methods: The transmission properties of figure-of-eight-shaped transformers were measured and compared to transformers with loop coils. Experiments to quantify the effect of decoupling from the B1 field of the MR system were conducted. Temperature measurements were performed to demonstrate the effective reduction of RF-induced heating. The transformers were investigated during active tracking experiments.Results: Coupling to the B1 field was reduced by 18 dB over conventional loop-shaped transformer coils. MR images showed a significantly reduced artifact for the figure-of-eight- shaped coils generated by local flip-angle amplification. Comparable transmission properties were seen for both transformer types. Temperature measurements showed a maximal temperature increase of 30K/3.5 K for an unsegmented/ segmented cable. With a segmented transmission line a robotic assistance system could be successfully localized using active tracking.Conclusion: The figure-of-eight-shaped transformer design reduces both RF field coupling with the MR system and artifact sizes. Anatomical structure close to the figure-of-eight-shaped transformer may be less obscured as with loop-shaped transformers if these transformers are integrated into e.g. intravascular catheters.

A Faraday effect position sensor for interventional magnetic resonance imaging

An optical sensor is presented which determines the position and one degree of orientation within a magnetic resonance tomograph. The sensor utilizes the Faraday effect to measure the local magnetic field, which is modulated by switching additional linear magnetic fields, the gradients. Existing methods for instrument localization during an interventional MR procedure often use electrically conducting structures at the instruments that can heat up excessively during MRI and are thus a significant danger for the patient. The proposed optical Faraday effect position sensor consists of non-magnetic and electrically non-conducting components only so that heating is avoided and the sensor could be applied safely even within the human body. With a non-magnetic prototype set-up, experiments were performed to demonstrate the possibility of measuring both the localization and the orientation in a magnetic resonance tomograph. In a 30 mT m(-1) gradient field, a localization uncertainty of 1.5 cm could be achieved.