Traditional poster

An MR thermometry dual-echo sequence is proposed here that offers advantages both in terms of temperature-to-noise ratio and image contrast, as compared to typically-used sequences. For thermometry in moving organs, the contrast properties of the proposed sequence allow blood vessels to be readily detected, for motion tracking purposes.

A fat temperature imaging technique based on multiple flip angle, multipoint Dixon acquisitions and a least square estimation scheme is proposed. Gradient recalled acquisition of 5 echo times with 3 different flip angles were obtained to separate the signals of methylene and methyl protons and to estimate T1's of these fatty acid species. Temperature images of a water-oil phantom were successfully obtained with previously obtained temperature coefficients demonstrating the feasibility of quantitative thermometry of fat. Since the acquisition time was 4-6 second, the technique seemed to be practical for temperature monitoring of fat-water tissues like breast under thermal therapies.

A novel body coil driven radiofrequency ablation (RFA) device is proposed. It provides an alternative to commercial available RFA device which required external power generator and large grounding pad. It allowed MR scanner as the sole modality to localize tumor, probe placement, RF power control, temperature mapping and tissue monitoring.

Multi-echo fat-water separation techniques, such as IDEAL, have been shown to be effective in measuring temperature changes in fatty tissue, but often make assumptions that allow them to linearize the model in order to simplify the computation of a solution. This can result in the addition of significant bias to the measurement of the temperature and the B0 field offset, both important parameters to monitor during therapeutic heat applications (tumor ablation, hyperthermia). In this work, the bias of a multi-peak IDEAL algorithm (without T2* decay) and a new nonlinear fitting algorithm is characterized using Monte Carlo simulations.

Multi-echo fat-water fitting techniques that separate the fat and water effects have been shown to be useful in measuring temperature in fat-water phantoms. In this study we explore optimization of echo time selection by minimizing the temperature noise using Cramer-Rao Lower Bound (CRLB) analysis. Accuracy of fitting is improved by including multiple fat peaks and T2* effects. Our approach finds the minimum temperature noise that has the minimum sensitivity to the values of nominally fixed parameters. The CRLB results were then confirmed in experiments with fat-water gelatin phantoms.

Decrease of the human brain temperature (1-2 oC in 15 minutes) was induced by intranasal cooling. The purpose of this study was to verify the cooling technique on the volunteers and to compare two methods for noninvasive monitoring of the relative brain temperature: MRSI with high spatial and reduced spectral resolution and PRF shift technique. Ability of the proposed brain cooling technique to induce moderate hypothermia was confirmed. Good agreement was found between relative temperatures measured by MRSI and PRF method. Both temperature mapping techniques can be used for monitoring the brain temperature changes during hypothermia.

In fat tissue, large susceptibility-related PRFS-based temperature errors can be expected, due to temporal changes in tissue susceptibility (χ) which lead to non-local magnetic field changes. This affects the PRF (hence, the measured temperature) of all water protons that experience this magnetic field change, leading to temperature errors. In order to conclusively assess the impact of temperature-induced χ changes on PRFS-based MRT, accurate and precise susceptibility measurements in human tissue are a prerequisite. We therefore measured dχ/dT of fat tissue of the human breast on a 14 T NMR spectrometer. A dependence of 0.0051 ppm/°C was found.

Susceptibility (χ) related field changes are commonly ignored in the application of proton resonance frequency shift (PRFS)-based MR thermometry (MRT) during thermal interventions, even though the temperature dependence of the χ of fat is in the same order of magnitude as the temperature dependence of the chemical shift of water. Its influence on PRFS-based MRT maps was investigated experimentally. The results showed that changes in χ _fat hamper the PRFS-based MRT method nonlocally. The measured errors were ranging between -4.6 °C and +4.1 °C. Important to stress is the fact that fat suppression is not a solution for this effect.

Safety and efficacy of tumor treatment using high intensity focus ultrasound requires accurate temperature measurement throughout the thermal procedure. In this work, we investigate how the noise in temperature measurements can be reduced by variations to this new ub-SSFP sequence.

Purpose was to investigate the dynamics of mild temperature induced contrast agent release from phosphatidylglyc-eroglycerol containing thermosensitive liposomes with encapsulated Gd-DTPA-BMA (Gd-TSL) in tumor tissue. Tumor bearing mice were investigated at 1.5T after intravenous injection. The temperature induced release of contrast agent at 42°C caused a fast and strong increase of T1-weighted signal. Immediately after i.v. injection heated tumor tissue was distinguishable from unheated tumor and muscle tissue. Unheated muscle tissue may thus be less affected by a potential anti tumor therapy based on TSL.

MR thermometry based on the proton resonance frequency (PRF) method (1) has gained good acceptance for guiding RF ablation of liver tumors (2). Major artifacts in the PRFS thermometry have recently been reported related to per-operatory changes of the tissue bulk susceptibility during RF heating (3). They are originating from gas bubbles formation, known as white cavitations’ artifacts in US imaging. We propose here a theoretical description of the effects and a first order correction that confirm the source of the spatially related errors in temperature maps and TD during power application.

MR thermometry in moving organs is a challenging application, as fairly subtle temperature-induced changes must be accurately measured in the presence of often much larger motion-induced changes. A novel approach at doing so is proposed here, which is both referenceless (does not require a baseline reference image) and user-independent.

MR thermometry is usually based on the temperature dependence of the proton resonance frequency (PRF), thus any magnetic field changes might be misinterpreted as temperature changes. Here we report on the effects of changes of susceptibility of surrounding air on the magnetic field inside an object. When the air was heated by 46 ºC, its susceptibility changed from χ _air = 3.6×10^-7 to χ _air = 2.7×10^-7, inducing an apparent additional temperature change of 1.9°C inside the object. For a more realistic surrounding air temperature increase of 10°C this could result in an error of 0.75°C.

A computationally efficient pipeline for 2D motion compensated PRF-thermometry and thermal dose measurements on moving abdominal organs is presented. The method is designed to address both, inter-scan and intra-scan artifacts by applying high frame-rate MRI coupled with a real-time image processing. The proposed MR-thermometry method was evaluated in both liver and kidney of 11 healthy volunteers and achieved a precision of less than 2 °C in 70 % of the pixels. The ability to perform MR-Thermometry and Dosimetry in-vivo was demonstrated on one HIFU-heating experiment on a porcine kidney.

A modified self-reference MR thermometry method is presented in this abstract. We circumvent the phase unwrapping procedure in the self-reference MR thermometry procedure by utilizing the phase gradient to estimate the baseline phase map. In the method proposed in this abstract, the phase map is modeled as a 2D polynomial. The components of the gradient of the model are then fitted to the components of the phase gradient using 2D weighted least squares. The proposed procedure is evaluated using two simulated MR thermometry data sets.

MRI is well suited to guide percutaneous interventions of liver lesions that are hardly visible with ultrasound or CT. Dedicated open MR systems are often used because they provide good patient access. This work presents first clinical experience with a new navigation solution that was used during RF ablation of liver tumors in a standard closed-bore scanner environment. After a special breathhold training, even double oblique access paths could be realized. RFA probe and thermally induced lesion could be reliably visualized with a VIBE sequence. While technical efforts are higher the times for needle placement and thermal ablation are comparable to those under CT guidance.

Model-based MR thermometry method based on the proton resonance frequency shift (PRFS) can effectively improve the temperature estimate accuracy of conventional phase different method. However, its temporal resolution need be improved for real-time temperature monitoring. To solve the problem, we applied highly accelerated PI to temperature mapping and used nonlocal regularization that extracts the prior from the acquired data themselves to stabilize the reconstruction. The method was demonstrated using whipping cream phantom. The results show that the nonlocal regularization can effectively increase the temporal resolution of PRFS while avoiding the introducing the bias to quantification, due to its data-driven property.

This study proposes the keyhole method in order to improve the time resolution of the proton resonance frequency (PRF) MR temperature monitoring technique. To evaluate proposed method, the values of Root Mean Square (RMS) error were compared with full phase encoded temperature images. And the paired t-test was performed for optimization at Keyhole technique. As a result of this study, >32 encoded images were reasonable in 95% confidence level whereas <32 encoded images showed statistically significant difference.

Convection Enhanced Delivery is a powerful method of delivering drugs to the CNS. We tested CED hardware and imaging methods by delivering Magnevist at three flow rates to the putamen in rhesus monkeys over a period of 7 days. T1-weighted images were acquired at three time points and from these the volume of distribution was characterized. Results demonstrate a compact distribution to the putamen with a flow rate-dependent volume.

Convection Enhanced Delivery is a powerful method of delivering drugs to the CNS. MR-visible tracers co-infused with drug will be useful to assess drug distribution. We tested four compounds (Magnevist and three Gd-labeled polylysines) with molecular weights between 0.5k Da and 200 kDa as potential tracers for CED. Compounds were tested in vitro to model CED parameters used to plan delivery into four rhesus monkeys. In vitro results demonstrated MW dependent CED distribution. In vivo results demonstrated distribution of the Magnevist in the putamen but little distribution of the polylysine as a result of binding and digestion of the polylysine.

A novel delivery system for MR guided precision minimally invasive access to deep brain structures is presented. A targeting accuracy of 0.5 mm at depths of 85-90mm is achieved in a phantom model. The delivery system is also used to insert cannula’s for infusing therapeutic agents via convection-enhanced delivery (CED). CED is demonstrated in a non-human primate with infusions to the thalamus, putamen and sub-thalamic nucleus. Accurate delivery of infusion cannula’s was achieved and CED infusions extending up to 10mm in diameter are demonstrated.

Osteochondritis Dissecans (OD) of the knee is a common articular lesion in adolescents and young adults. Retrograde drilling is an alternative in surgical treatment before more invasive and complex procedures are necessary; however drilling under fluoroscopic guidance via the epiphyses is technically challenging in terms of maintaining drill depth and accuracy of drill placement. The purpose of this study was to evaluate applicability and accuracy of a passive navigation method by drilling a simulated OD target at the knee with MRI guidance. Interactive MR navigation allowed precise drilling of OCD of the knee. Targeting was possible with a single drilling.

During an MR-guided breast intervention, placement of a needle or probe requires accurate localization of the target. To allow rapid identification of these errors, avoid unnecessary trauma to the patient, and minimize scanner time, a real-time MR acquisition and display was implemented to allow the physician to monitor both the progress of the insertion and the procedure itself, such as core needle biopsy or tumor ablation. Flexibility was added to adjust the scan plane in real-time and make tradeoffs between update display rate and image quality.

Percutaneous interventions involve the navigation of a needle or probe to a target location. MRI is well-suited to guide these procedures as it offers good soft tissue target visualization and no ionizing radiation. High field wide bore MRI has stimulated interest in performing more of these procedures, but workflow and procedure speed are significant hurdles for full adoption. Here, a navigation tool for guiding and tracking the needle in real-time under MRI is presented.

MRI has been shown to be of great clinical utility for the guidance of various procedures. In a closed-bore scanner, the simplest approach is to manipulate the instrument outside the bore and move the patient inside for control imaging only. Potential benefits for percutaneous biopsies in the liver have been investigated in 15 patients by using a flexible add-on navigation solution which even allowed interventions in obese patients. Real time navigation was provided by following the virtual instrument on properly reformatted images of a 3D roadmap. In combination with a specific breathhold protocol, punctures could be reliably performed in reasonable times.

Purpose: To test the feasibility and accuracy of a tool that allows for interactive adjustments of the needle plane during a MR guided puncture Method: Experiments were performed in vivo and in phantoms using a gradient based navigation system for real time MR guided punctures in a wide bore MR imager. To assess for accuracy of the system the distance of the needle tip (virtual and real) to the target was determined on MR control scans. Result: The mean 3D total error was 4.9 ±2.8mm in the phantom. The system error was less than 2 mm. In the animal, successful punctures of the target structures could be confirmed in all punctures. Conclusion: The combination of image overlay with real time adjustment of the virtual needle and real-time imaging feedback provides an accurate and intuitive means to perform percutaneous interventions in a wide bore MR imager.

Real time MR-guided prostate biopsy at 3 Tesla in animal experiment proved to be a fast an precise method to perform prostate biopsies.

As accurate needle positioning helps the prostate cancer detection and treatment, a number of MRI-compatible robots have been introduced. However, problems exist due to the strong magnetic field and limited workspace. Pneumatic actuator has the minimum distraction in the environment. However, it has poor controllability. To overcome the controllability problem, a simple external damping mechanism that can enhance accuracy was developed. Based on the actuator mechanism and workflow optimized modular design approaches, a new pneumatically actuated 4-DOF parallel robot for MRI-guided prostate intervention was developed. A preliminary evaluation was conducted with satisfying actuator average position error of 0.2mm.

This paper reports the design, development and MRI compatibility evaluation of a transrectal prostate robot for MRI-guided intervention. The robot employs an automated needle guide with the goal of increasing needle placement accuracy and reducing interventional procedure times. The design of the robot, employing piezo-ceramic-motor actuated needle guide and manual needle insertion, is reported. Results of a MRI compatibility study show no reduction of MRI image SNR with the motors disabled and a 40% to 60% reduction with the motors enabled. The addition of RF shielding is shown to significantly reduce SNR degradation to the presence of the robotic device.

A limb-positioning mechanical platform device was developed for remote orientation of the arm to make use of the magic angle effect for imaging tendons. The platform is MR-compatible and actuated by rotary air-driven motors. Clinical trials are imminent.

MRI permits immediate depiction of ablated tissue zones for intra-procedural monitoring during irriversible electroporation (IRE) ablation procedures. MRI monitoring offers the potential to permit intra-procedural optimization of IRE procedures to ensure complete ablation of targeted tissue volumes.

Electroporation involves targeted delivery of electrical pulses to permeabilize cell membranes, either reversible or irreversible. Irreversible electroporation (IRE), as a new tissue ablation technique, induces tissue necrosis due to permanent cell membrane defects. Assessment of tissue response to IRE may be critical. For our study, we demonstrate that inversion recovery prepared contrast enhancement imaging, with TI adjusted to null the signal intensity from the reversible zone, can visualize the IRE ablated tissue margin (differentiating reversible/irreversible zones) to provide an accurate prediction of ablation. This technique can early detect tissue response to IRE and might be helpful to guide further treatments.

The use of antibodies to target toxic amyloid-beta peptides (Aβ) in the brain of Alzheimer’s patients has shown promise in clinical trials but still faces some difficulties. The blood-brain barrier remains a major obstacle; preventing intravenously delivered antibodies from reaching the brain. In this study, we use transcranial MRI-guided focused ultrasound to efficiently deliver antibodies to the brain of a mouse model of Alzheimer’s disease and evaluate the efficacy of this treatment. We found that delivery of the antibody is localized to targeted regions and yields a rapid and significant reduction of Aβ plaque load from a single treatment.

Four-dimensional (4D) transcatheter intra-arterial perfusion (TRIP) MRI is a monitoring technique that involves targeted intra-arterial (IA) injections of gadolinium (Gd) (<0.001 mmol/kg) to help quantify volumetric tissue-perfusion changes over time. This technique can be used to objectively monitor tumor perfusion changes after UAE in VX2 rabbits. The technique may have future clinical application in optimizing endpoints during UAE.