Accuracy of 3D real-time MRI temperature mapping in gel phantoms during microwave heating.

BACKGROUND: Interventional magnetic resonance imaging (MRI) can provide a comprehensive setting for microwave ablation of tumors with real-time monitoring of the energy delivery using MRI-based temperature mapping. The purpose of this study was to quantify the accuracy of three-dimensional (3D) real-time MRI temperature mapping during microwave heating in vitro by comparing MRI thermometry data to reference data measured by fiber-optical thermometry. METHODS: Nine phantom experiments were evaluated in agar-based gel phantoms using an in-room MR-conditional microwave system and MRI thermometry. MRI measurements were performed for 700 s (25 slices; temporal resolution 2 s). The temperature was monitored with two fiber-optical temperature sensors approximately 5 mm and 10 mm distant from the microwave antenna. Temperature curves of the sensors were compared to MRI temperature data of single-voxel regions of interest (ROIs) at the sensor tips; the accuracy of MRI thermometry was assessed as the root-mean-squared (RMS)-averaged temperature difference. Eighteen neighboring voxels around the original ROI were also evaluated and the voxel with the smallest temperature difference was additionally selected for further evaluation. RESULTS: The maximum temperature changes measured by the fiber-optical sensors ranged from 7.3 K to 50.7 K. The median RMS-averaged temperature differences in the originally selected voxels ranged from 1.4 K to 3.4 K. When evaluating the minimum-difference voxel from the neighborhood, the temperature differences ranged from 0.5 K to 0.9 K. The microwave antenna and the MRI-conditional in-room microwave generator did not induce relevant radiofrequency artifacts. CONCLUSION: Accurate 3D real-time MRI temperature mapping during microwave heating with very low RMS-averaged temperature errors below 1 K is feasible in gel phantoms. RELEVANCE STATEMENT: Accurate MRI-based volumetric real-time monitoring of temperature distribution and thermal dose is highly relevant in clinical MRI-based interventions and can be expected to improve local tumor control, as well as procedural safety by extending the limits of thermal (e.g., microwave) ablation of tumors in the liver and in other organs. KEY POINTS: Interventional MRI can provide a comprehensive setting for the microwave ablation of tumors. MRI can monitor the microwave ablation using real-time MRI-based temperature mapping. 3D real-time MRI temperature mapping during microwave heating is feasible. Measured temperature errors were below 1 °C in gel phantoms. The active in-room microwave generator did not induce any relevant radiofrequency artifacts.

3D motion strategy for online volumetric thermometry using simultaneous multi-slice EPI at 1.5T: an evaluation study.

PURPOSE: In presence of respiratory motion, temperature mapping is altered by in-plane and through-plane displacements between successive acquisitions together with periodic phase variations. Fast 2D Echo Planar Imaging (EPI) sequence can accommodate intra-scan motion, but limited volume coverage and inter-scan motion remain a challenge during free-breathing acquisition since position offsets can arise between the different slices. METHOD: To address this limitation, we evaluated a 2D simultaneous multi-slice EPI sequence with multiband (MB) acceleration during radiofrequency ablation on a mobile gel and in the liver of a volunteer (no heating). The sequence was evaluated in terms of resulting inter-scan motion, temperature uncertainty and elevation, potential false-positive heating and repeatability. Lastly, to account for potential through-plane motion, a 3D motion compensation pipeline was implemented and evaluated. RESULTS: In-plane motion was compensated whatever the MB factor and temperature distribution was found in agreement during both the heating and cooling periods. No obvious false-positive temperature was observed under the conditions being investigated. Repeatability of measurements results in a 95% uncertainty below 2 °C for MB1 and MB2. Uncertainty up to 4.5 °C was reported with MB3 together with the presence of aliasing artifacts. Lastly, fast simultaneous multi-slice EPI combined with 3D motion compensation reduce residual out-of-plane motion. CONCLUSION: Volumetric temperature imaging (12 slices/700 ms) could be performed with 2 °C accuracy or less, and offer tradeoffs in acquisition time or volume coverage. Such a strategy is expected to increase procedure safety by monitoring large volumes more rapidly for MR-guided thermotherapy on mobile organs.

Real-time automatic temperature regulation during in vivo MRI-guided laser-induced thermotherapy (MR-LITT).

Precise control of tissue temperature during Laser-Induced Thermotherapy (LITT) procedures has the potential to improve the clinical efficiency and safety of such minimally invasive therapies. We present a method to automatically regulate in vivo the temperature increase during LITT using real-time rapid volumetric Magnetic Resonance thermometry (8 slices acquired every second, with an in-plane resolution of 1.4 mmx1.4 mm and a slice thickness of 3 mm) using the proton-resonance frequency (PRF) shift technique. The laser output power is adjusted every second using a feedback control algorithm (proportional-integral-derivative controller) to force maximal tissue temperature in the targeted region to follow a predefined temperature-time profile. The root-mean-square of the difference between the target temperature and the measured temperature ranged between 0.5 °C and 1.4 °C, for temperature increases between + 5 °C to + 30 °C above body temperature and a long heating duration (up to 15 min), showing excellent accuracy and stability of the method. These results were obtained on a 1.5 T clinical MRI scanner, showing a potential immediate clinical application of such a temperature controller during MR-guided LITT.

MRI monitoring of temperature and displacement for transcranial focus ultrasound applications.

BACKGROUND: Transcranial focus ultrasound applications applied under MRI-guidance benefit from unrivaled monitoring capabilities, allowing the recording of real-time anatomical information and biomarkers like the temperature rise and/or displacement induced by the acoustic radiation force. Having both of these measurements could allow for better targeting of brain structures, with improved therapy monitoring and safety. METHOD: We investigated the use of a novel MRI-pulse sequence described previously in Bour et al., (2017) to quantify both the displacement and temperature changes under various ultrasound sonication conditions and in different regions of the brain. The method was evaluated in vivo in a non-human primate under anesthesia using a single-element transducer (f = 850 kHz) in a setting that could mimic clinical applications. Acquisition was performed at 3 T on a clinical imaging system using a modified single-shot gradient echo EPI sequence integrating a bipolar motion-sensitive encoding gradient. Four slices were acquired sequentially perpendicularly or axially to the direction of the ultrasound beam with a 1-Hz update frequency and an isotropic spatial resolution of 2-mm. A total of twenty-four acquisitions were performed in three different sets of experiments. Measurement uncertainty of the sequence was investigated under different acoustic power deposition and in different regions of the brain. Acoustic simulation and thermal modeling were performed and compared to experimental data. RESULTS: The sequence simultaneously provides relevant information about the focal spot location and visualization of heating of brain structures: 1) The sequence localized the acoustic focus both along as well as perpendicular to the ultrasound direction. Tissue displacements ranged from 1 to 2 µm. 2) Thermal rise was only observed at the vicinity of the skull. Temperature increase ranged between 1 and 2 °C and was observed delayed relative the sonication due to thermal diffusion. 3) The fast frame rate imaging was able to highlight magnetic susceptibility artifacts related to breathing, for the most caudal slices. We demonstrated that respiratory triggering successfully restored the sensitivity of the method (from 0.7 µm to 0.2 µm). 4) These results were corroborated by acoustic simulations. CONCLUSIONS: The current rapid, multi-slice acquisition and real-time implementation of temperature and displacement visualization may be useful in clinical practices. It may help defining operational safety margins, improving therapy precision and efficacy. Simulations were in good agreement with experimental data and may thus be used prior treatment for procedure planning.

MR-ARFI-based method for the quantitative measurement of tissue elasticity: application for monitoring HIFU therapy.

Monitoring thermal therapies through medical imaging is essential in order to ensure that they are safe, efficient and reliable. In this paper, we propose a new approach, halfway between MR acoustic radiation force imaging (MR-ARFI) and MR elastography (MRE), allowing for the quantitative measurement of the elastic modulus of tissue in a highly localized manner. It relies on the simulation of the MR-ARFI profile, which depends on tissue biomechanical properties, and on the identification of tissue elasticity through the fitting of experimental displacement images measured using rapid MR-ARFI. This method was specifically developed to monitor MR-guided high intensity focused ultrasound (MRgHIFU) therapy. Elasticity changes were followed during HIFU ablations (N = 6) performed ex vivo in porcine muscle samples, and were compared to temperature changes measured by MR-thermometry. Shear modulus was found to increase consistently and steadily a few seconds after the heating started, and such changes were found to be irreversible. The shear modulus was found to increase from 1.49 ± 0.48 kPa (before ablation) to 3.69 ± 0.93 kPa (after ablation and cooling). Thanks to its ability to perform quantitative elasticity measurements in a highly localized manner around the focal spot, this method proved to be particularly attractive for monitoring HIFU ablations.

Real-time 3D ultrasound based motion tracking for the treatment of mobile organs with MR-guided high-intensity focused ultrasound.

INTRODUCTION: Magnetic resonance-guided high-intensity focused ultrasound (MRgHIFU) treatments of mobile organs require locking the HIFU beam on the targeted tissue to maximise heating efficiency. We propose to use a standalone 3 D ultrasound (US)-based motion correction technique using the HIFU transducer in pulse-echo mode. Validation of the method was performed in vitro and in vivo in the liver of pig under MR-thermometry. METHODS: 3 D-motion estimation was implemented using ultrasonic speckle-tracking between consecutive acquisitions. Displacement was estimated along four sub-apertures of the HIFU transducer by computing the normalised cross-correlation of backscattered signals followed by a triangulation algorithm. The HIFU beam was steered accordingly and energy was delivered under real-time MR-thermometry (using the proton resonance frequency shift method with online motion compensation and correction of associated susceptibility artefacts). An MR-navigator echo was used to assess the quality of the US-based motion correction. RESULTS: Displacement estimations from US measurements were in good agreement with 1 D MR-navigator echo readings. In vitro, the maximum temperature increase was improved by 37% as compared to experiments performed without motion correction and temperature distribution remained much more focussed. Similar results were reported in vivo, with an increase of 35% on the maximum temperature using this US-based HIFU target locking. CONCLUSION: This standalone 3D US-based motion correction technique is robust and allows maintaining the HIFU focal spot in the presence of motion without adding any burden or complexity to MR thermal imaging. In vitro and in vivo results showed about 35% improvement in heating efficiency when focus position was locked on the target using the proposed technique.

Real-time monitoring of tissue displacement and temperature changes during MR-guided high intensity focused ultrasound.

PURPOSE: The therapy endpoint most commonly used in MR-guided high intensity focused ultrasound is the thermal dose. Although namely correlated with nonviable tissue, it does not account for changes in mechanical properties of tissue during ablation. This study presents a new acquisition sequence for multislice, subsecond and simultaneous imaging of tissue temperature and displacement during ablation. METHODS: A single-shot echo planar imaging sequence was implemented using a pair of motion-encoding gradients, with alternated polarities. A first ultrasound pulse was synchronized on the second lobe of the motion-encoding gradients and followed by continuous sonication to induce a local temperature increase in ex vivo muscle and in vivo on pig liver. Lastly, the method was evaluated in the brain of two volunteers to assess method's precision. RESULTS: For thermal doses higher than the lethal threshold, displacement amplitude was reduced by 21% and 28% at the focal point in muscle and liver, respectively. Displacement value remained nearly constant for nonlethal thermal doses values. The mean standard deviation of temperature and displacement in the brain of volunteers remained below 0.8 °C and 2.5 µm. CONCLUSION: This new fast imaging sequence provides real-time measurement of temperature distribution and displacement at the focus during HIFU ablation. Magn Reson Med 78:1911-1921, 2017. © 2017 International Society for Magnetic Resonance in Medicine.

Improved cardiac magnetic resonance thermometry and dosimetry for monitoring lesion formation during catheter ablation.

PURPOSE: A new real-time MR-thermometry pipeline was developed to measure multiple temperature images per heartbeat with 1.6×1.6×3 mm3 spatial resolution. The method was evaluated on 10 healthy volunteers and during radiofrequency ablation (RFA) in sheep. METHODS: Multislice, electrocardiogram-triggered, echo-planar imaging was combined with parallel imaging, under free breathing conditions. In-plane respiratory motion was corrected on magnitude images by an optical flow algorithm. Motion-related susceptibility artifacts were compensated on phase images by an algorithm based on Principal Component Analysis. Correction of phase drift and temporal filter were included in the pipeline implemented in the Gadgetron framework. Contact electrograms were recorded simultaneously with MR thermometry by an MR-compatible ablation catheter. RESULTS: The temporal standard deviation of temperature in the left ventricle remained below 2 °C on each volunteer. In sheep, focal heated regions near the catheter tip were observed on temperature images (maximal temperature increase of 38 °C) during RFA, with contact electrograms of acceptable quality. Thermal lesion dimensions at gross pathology were in agreement with those observed on thermal dose images. CONCLUSION: This fully automated MR thermometry pipeline (five images/heartbeat) provides direct assessment of lesion formation in the heart during catheter-based RFA, which may improve treatment of cardiac arrhythmia by ablation. Magn Reson Med 77:673-683, 2017. © 2016 International Society for Magnetic Resonance in Medicine.

Magnetic resonance-compatible model of isolated working heart from large animal for multimodal assessment of cardiac function, electrophysiology, and metabolism.

To provide a model close to the human heart, and to study intrinsic cardiac function at the same time as electromechanical coupling, we developed a magnetic resonance (MR)-compatible setup of isolated working perfused pig hearts. Hearts from pigs (40 kg, n = 20) and sheep (n = 1) were blood perfused ex vivo in the working mode with and without loaded right ventricle (RV), for 80 min. Cardiac function was assessed by measuring left intraventricular pressure and left ventricular (LV) ejection fraction (LVEF), aortic and mitral valve dynamics, and native T1 mapping with MR imaging (1.5 Tesla). Potential myocardial alterations were assessed at the end of ex vivo perfusion from late-Gadolinium enhancement T1 mapping. The ex vivo cardiac function was stable across the 80 min of perfusion. Aortic flow and LV-dP/dtmin were significantly higher (P < 0.05) in hearts perfused with loaded RV, without differences for heart rate, maximal and minimal LV pressure, LV-dP/dtmax, LVEF, and kinetics of aortic and mitral valves. T1 mapping analysis showed a spatially homogeneous distribution over the LV. Simultaneous recording of hemodynamics, LVEF, and local cardiac electrophysiological signals were then successfully performed at baseline and during electrical pacing protocols without inducing alteration of MR images. Finally, (31)P nuclear MR spectroscopy (9.4 T) was also performed in two pig hearts, showing phosphocreatine-to-ATP ratio in accordance with data previously reported in vivo. We demonstrate the feasibility to perfuse isolated pig hearts in the working mode, inside an MR environment, allowing simultaneous assessment of cardiac structure, mechanics, and electrophysiology, illustrating examples of potential applications.