Juan M. Santos

Compressed Sensing MRI

This article reviews the requirements for successful compressed sensing (CS), describes their natural fit to MRI, and gives examples of four interesting applications of CS in MRI. The authors emphasize on an intuitive understanding of CS by describing the CS reconstruction as a process of interference cancellation. There is also an emphasis on the understanding of the driving factors in applications, including limitations imposed by MRI hardware, by the characteristics of different types of images, and by clinical concerns.

PROMO: Real-time prospective motion correction in MRI using image-based tracking.

Artifacts caused by patient motion during scanning remain a serious problem in most MRI applications. The prospective motion correction technique attempts to address this problem at its source by keeping the measurement coordinate system fixed with respect to the patient throughout the entire scan process. In this study, a new image‐based approach for prospective motion correction is described, which utilizes three orthogonal two‐dimensional spiral navigator acquisitions, along with a flexible image‐based tracking method based on the extended Kalman filter algorithm for online motion measurement. The spiral navigator/extended Kalman filter framework offers the advantages of image‐domain tracking within patient‐specific regions‐of‐interest and reduced sensitivity to off‐resonance‐induced corruption of rigid‐body motion estimates. The performance of the method was tested using offline computer simulations and online in vivo head motion experiments. In vivo validation results covering a broad range of staged head motions indicate a steady‐state error of less than 10% of the motion magnitude, even for large compound motions that included rotations over 15 deg. A preliminary in vivo application in three‐dimensional inversion recovery spoiled gradient echo (IR‐SPGR) and three‐dimensional fast spin echo (FSE) sequences demonstrates the effectiveness of the spiral navigator/extended Kalman filter framework for correcting three‐dimensional rigid‐body head motion artifacts prospectively in high‐resolution three‐dimensional MRI scans. Magn Reson Med, 2010.

Using Real-Time MRI to Quantify Altered Joint Kinematics in Subjects with Patellofemoral Pain and to Evaluate the Effects of a Patellar Brace or Sleeve on Joint Motion

Abnormal patellofemoral joint motion is a possible cause of patellofemoral pain, and patellar braces are thought to alleviate pain by restoring normal joint kinematics. We evaluated whether females with patellofemoral pain exhibit abnormal patellofemoral joint kinematics during dynamic, weight‐bearing knee extension and assessed the effects of knee braces on patellofemoral motion. Real‐time magnetic resonance (MR) images of the patellofemoral joints of 36 female volunteers (13 pain‐free controls, 23 patellofemoral pain) were acquired during weight‐bearing knee extension. Pain subjects were also imaged while wearing a patellar‐stabilizing brace and a patellar sleeve. We measured axial‐plane kinematics from the images. Females with patellofemoral pain exhibited increased lateral translation of the patella for knee flexion angles between 0°and 50° (p = 0.03), and increased lateral tilt for knee flexion angles between 0° and 20° (p = 0.04). The brace and sleeve reduced the lateral translation of the patella; however, the brace reduced lateral displacement more than the sleeve (p = 0.006). The brace reduced patellar tilt near full extension (p = 0.001), while the sleeve had no effect on patellar tilt. Our results indicate that some subjects with patellofemoral pain exhibit abnormal weight‐bearing joint kinematics and that braces may be effective in reducing patellar maltracking in these subjects. Published by Wiley Periodicals, Inc. J Orthop Res 27: 571–577, 2009

Real‐time cardiac MRI at 3 tesla

Real‐time cardiac and coronary MRI at 1.5T is relatively “signal starved” and the 3T platform is attractive for its immediate factor of two increase in magnetization. Cardiac imaging at 3T, however, is both subtly and significantly different from imaging at 1.5T because of increased susceptibility artifacts, differences in tissue relaxation, and RF homogeneity issues. New RF excitation and pulse sequence designs are presented which deal with the fat‐suppression requirements and off‐resonance issues at 3T. Real‐time cardiac imaging at 3T is demonstrated with high blood SNR, blood‐myocardium CNR, resolution, and image quality, using new spectral‐spatial RF pulses and fast spiral gradient echo pulse sequences. The proposed sequence achieves 1.5 mm in‐plane resolution over a 20 cm FOV, with a 5.52 mm measured slice thickness and 32 dB of lipid suppression. Complete images are acquired every 120 ms and are reconstructed and displayed at 24 frames/sec using a sliding window. Results from healthy volunteers show improved image quality, a 53% improvement in blood SNR efficiency, and a 232% improvement in blood‐myocardium CNR efficiency compared to 1.5T. Magn Reson Med 51:655–660, 2004.

Flexible real-time magnetic resonance imaging framework

The extension of MR imaging to new applications has demonstrated the limitations of the architecture of current real-time systems. Traditional real-time implementations provide continuous acquisition of data and modification of basic sequence parameters on the fly. We have extended the concept of real-time MRI by designing a system that drives the examinations from a real-time localizer and then gets reconfigured for different imaging modes. Upon operator request or automatic feedback the system can immediately generate a new pulse sequence or change fundamental aspects of the acquisition such as gradient waveforms excitation pulses and scan planes. This framework has been implemented by connecting a data processing and control workstation to a conventional clinical scanner. Key components on the design of this framework are the data communication and control mechanisms, reconstruction algorithms optimized for real-time and adaptability, flexible user interface and extensible user interaction. In this paper we describe the various components that comprise this system. Some of the applications implemented in this framework include real-time catheter tracking embedded in high frame rate real-time imaging and immediate switching between real-time localizer and high-resolution volume imaging for coronary angiography applications.

Hybrid referenceless and multibaseline subtraction MR thermometry for monitoring thermal therapies in moving organs.

PURPOSE Magnetic resonance thermometry using the proton resonance frequency (PRF) shift is a promising technique for guiding thermal ablation. For temperature monitoring in moving organs, such as the liver and the heart, problems with motion must be addressed. Multi-baseline subtraction techniques have been proposed, which use a library of baseline images covering the respiratory and cardiac cycle. However, main field shifts due to lung and diaphragm motion can cause large inaccuracies in multi-baseline subtraction. Referenceless thermometry methods based on polynomial phase regression are immune to motion and susceptibility shifts. While referenceless methods can accurately estimate temperature within the organ, in general, the background phase at organ/tissue interfaces requires large polynomial orders to fit, leading to increased danger that the heated region itself will be fitted by the polynomial and thermal dose will be underestimated. In this paper, a hybrid method for PRF thermometry in moving organs is presented that combines the strengths of referenceless and multi-baseline thermometry. METHODS The hybrid image model assumes that three sources contribute to image phase during thermal treatment: Background anatomical phase, spatially smooth phase deviations, and focal, heat-induced phase shifts. The new model and temperature estimation algorithm were tested in the heart and liver of normal volunteers, in a moving phantom HIFU heating experiment, and in numerical simulations of thermal ablation. The results were compared to multi-baseline and referenceless methods alone. RESULTS The hybrid method allows for in vivo temperature estimation in the liver and the heart with lower temperature uncertainty compared to multi-baseline and referenceless methods. The moving phantom HIFU experiment showed that the method accurately estimates temperature during motion in the presence of smooth main field shifts. Numerical simulations illustrated the methods sensitivity to algorithm parameters and hot spot features. CONCLUSIONS This new hybrid method for MR thermometry in moving organs combines the strengths of both multi-baseline subtraction and referenceless thermometry and overcomes their fundamental weaknesses.

Real‐time MR thermometry for monitoring HIFU ablations of the liver

A high‐resolution and high‐speed pulse sequence is presented for monitoring high‐intensity focused ultrasound ablations in the liver in the presence of motion. The sequence utilizes polynomial‐order phase saturation bands to perform outer volume suppression, followed by spatial‐spectral excitation and three readout segmented echo‐planar imaging interleaves. Images are processed with referenceless thermometry to create temperature‐rise images every frame. The sequence and reconstruction were implemented in RTHawk and used to image stationary and moving sonications in a polyacrylamide gel phantom (62.4 acoustic W, 50 sec, 550 kHz). Temperature‐rise images were compared between moving and stationary experiments. Heating spots and corresponding temperature‐rise plots matched very well. The stationary sonication had a temperature standard deviation of 0.15° C compared to values of 0.28° C and 0.43° C measured for two manually moved sonications at different velocities. Moving the phantom (while not heating) with respect to the transducer did not cause false temperature rises, despite susceptibility changes. The system was tested on nonheated livers of five normal volunteers. The mean temperature rise was − 0.05° C, with a standard deviation of 1.48° C. This standard deviation is acceptable for monitoring high‐intensity focused ultrasound ablations, suggesting real‐time imaging of moving high‐intensity focused ultrasound sonications can be clinically possible. Magn Reson Med, 2010.

Single breath-hold whole-heart MRA using variable-density spirals at 3t

Multislice breath‐held coronary imaging techniques conventionally lack the coverage of free‐breathing 3D acquisitions but use a considerably shorter acquisition window during the cardiac cycle. This produces images with significantly less motion artifact but a lower signal‐to‐noise ratio (SNR). By using the extra SNR available at 3 T and undersampling k‐space without introducing significant aliasing artifacts, we were able to acquire high‐resolution fat‐suppressed images of the whole heart in 17 heartbeats (a single breath‐hold). The basic pulse sequence consists of a spectral‐spatial excitation followed by a variable‐density spiral readout. This is combined with real‐time localization and a real‐time prospective shim correction. Images are reconstructed with the use of gridding, and advanced techniques are used to reduce aliasing artifacts. Magn Reson Med, 2006.

Differences in patellofemoral kinematics between weight-bearing and non-weight-bearing conditions in patients with patellofemoral pain

Patellar maltracking is thought to be one source of patellofemoral pain. Measurements of patellar tracking are frequently obtained during non‐weight‐bearing knee extension; however, pain typically arises during highly loaded activities, such as squatting, stair climbing, and running. It is unclear whether patellofemoral joint kinematics during lightly loaded tasks replicate patellofemoral joint motion during weight‐bearing activities. The purpose of this study was to: evaluate differences between upright, weight‐bearing and supine, non‐weight‐bearing joint kinematics in patients with patellofemoral pain; and evaluate whether the kinematics in subjects with maltracking respond differently to weight‐bearing than those in nonmaltrackers. We used real‐time magnetic resonance imaging to visualize the patellofemoral joint during dynamic knee extension from 30° to 0° of knee flexion during two conditions: upright, weight‐bearing and supine, non‐weight‐bearing. We compared patellofemoral kinematics measured from the images. The patella translated more laterally during the supine task compared to the weight‐bearing task for knee flexion angles between 0° and 5° (p = 0.001). The kinematics of the maltrackers responded differently to joint loading than those of the non‐maltrackers. In subjects with excessive lateral patellar translation, the patella translated more laterally during upright, weight‐bearing knee extension for knee flexion angles between 25° and 30° (p = 0.001). However, in subjects with normal patellar translation, the patella translated more laterally during supine, non‐weight‐bearing knee extension near full extension (p = 0.001). These results suggest that patellofemoral kinematics measured during supine, unloaded tasks do not accurately represent the joint motion during weight‐bearing activities.

Helical MR: continuously moving table axial imaging with radial acquisitions.

A technique for extended field of view MRI is presented. Similar to helical computed tomography, the method utilizes a continuously moving patient table, a 2D axial slice that remains fixed relative to the MRI magnet, and a radial k‐space trajectory. A fully refocused SSFP acquisition enables spatial resolution comparable to current clinical protocols in scan times that are sufficiently short to allow a reasonable breathhold duration. RF transmission and signal reception are performed using the RF body coil and the images are reconstructed in real time. Experimental results are presented that illustrate the techniques ability to resolve small structures in the table‐motion direction. Simulation experiments to study the steady‐state response of the fully refocused SSFP acquisition during continuous table motion are also presented. Finally, whole body images of healthy volunteers demonstrate the high image quality achieved using the helical MRI approach. Magn Reson Med 50:1053–1060, 2003.