A promising 3-D spiral-based MR thermometry technique can help guide focused ultrasound (US) treatment for neuro soft tissue ablation, providing more accuracy and faster procedure times than the standard 2-D technique, according to new research.
“We anticipate that spiral-based 3-D MR thermometry will have applications for temperature monitoring of any static brain or extra-cranial soft tissue ablation procedure,” said Matthew Geeslin, MD, MS, a radiology fellow at the University of Virginia Health System, in Charlottesville, whose project was funded through a 2014 Toshiba America Medical Systems/RSNA Research Resident Grant.
The current clinical standard for planning and monitoring MR-guided focused ultrasound (MRgFUS) for brain applications — 2-D MR thermometry — has limitations including accuracy of temperature mapping, monitoring near-field heating and lengthy procedure time. In his research, Dr. Geeslin and his advisor sought evaluate a spiral-based 3-D MR thermometry technique that addresses those limitations and improves on the standard technique.
All MR acquisition techniques are subject to artifacts induced by off-resonance, Dr. Geeslin explained. Typical methods for acquiring thermal image maps, such as echo-planar imaging or spin-warp imaging, rely on a Cartesian trajectory, which is subject to geometric distortions.
Spiral imaging offers an advantage over echo-planar imaging because off-resonance does not shift the position of the “hot spot,” Dr. Geeslin said.
“Image blur from off-resonance effects can be a concern with spiral scanning, but our lab group had already developed a reliable set of image construction methods for minimizing off-resonance effects from both main field inhomogeneity and concomitant gradient fields,” he said.
There are many ways to sample the data in k-space, including spiral, Cartesian, radial and zig-zag acquisitions. Each sampling method has pitfalls related to magnetic field inhomogeneity, Dr. Geeslin said. In spiral acquisitions, off-resonance from field inhomogeneity manifests as image blur.
“The techniques for reducing image blur are particularly beneficial in spiral-based MR thermometry because spiral, rather than Cartesian, sampling of k-space trades shifting of the hot spot for image blur,” he said. “This has safety advantages during both procedure planning and completion.”
Under the guidance of project advisor Max Wintermark, MD, MS, a professor of radiology and chief of neuroradiology at Stanford University Medical Center, in Palo Alto, CA, Dr. Geeslin and colleagues evaluated their techniques in the brains of pigs. They began by performing focused sonication in two phases in order to gather as much thermometry data as possible from each animal.
The first phase of sonication was non-ablative, depositing energy high enough to cause a temperature rise but low enough to avoid forming a durable lesion. The technique used both high- and low-power sonication parameters to generate non-ablative temperature profiles.
“This also increased the number of data points for comparison of 2-D and 3-D techniques and helped evaluate our 3-D spiral sequence over a range of temperatures,” Dr. Geeslin said.
The second phase aimed at creating the type of lesion formed during a focused US procedure for essential tremor.
“Essential tremor was our most immediate clinical target, given the then-ongoing focused ultrasound trial and the availability of an established animal model for brain applications,” Dr. Geeslin said, noting that MR-guided focused ultrasound treatment of essential tremor is now approved by the U.S. Food and Drug Administration (FDA). The researchers then used both a commercial 2-D thermometry sequence and their own spiral-based 3-D sequence to generate temperature maps.
The researchers determined that while the 3-D temperature maps displayed greater noise than the 2-D maps, the 3-D maps did not suffer from distortion. In addition, the availability of a new, smaller surface coil provided a better signal-to-noise ratio when compared to their usual body coil, improving the temperature resolution.
Ultimately, Dr. Geeslin concluded that a single spiral-based 3-D acquisition could lead to “reduced procedure time and a more comfortable experience for the patient.”
Data from Dr. Geeslin’s RSNA research led to a 2015 grant from the Wallace Coulter Foundation, which the research group is using to continue investigating the 3-D spiral-based MR thermometry technique.
“The next phase involves implementing the use of Kalman filtering to further improve the accuracy of temperature and thermal dose calculation in spiral-based MR thermometry,” he said.
Dr. Wintermark, chair of RSNA’s Public Information Committee, noted that supporting research for residents like Dr. Geeslin often has a significant impact on their pursuit of academic careers.
“These researchers also become role models for their co-residents,” Dr. Wintermark said. “RSNA grants like the Resident Research Grant foster a research and academically-oriented culture in radiology departments.”
In terms of pursuing research, Dr. Geeslin says that it is important to choose a topic that will be a sustaining source of motivation.
“Additionally, it is helpful to both your field and your future funding prospects to be engaged in research that progressively builds on itself through your training and early career,” Dr. Geeslin said. “And an excellent and experienced mentor, like Dr. Wintermark, can be a window into your future.”
Matthew Geeslin, MD, MS
2014 Toshiba America Medical Systems/RSNA Research Resident Grant.
“Evaluation of Spiral-based 3-D MR Thermometry for Brain Applications of MR-guided Focused Ultrasound in a Porcine Model.”
“Prior to notification of the RSNA research award, I considered myself committed to a career as a physician scientist. Assuming the role of primary investigator in this research project absolutely strengthened my resolve to pursue such a career, but more importantly, perhaps, it significantly increases the likelihood that I will succeed in such a capacity.”
“Our technique eliminates one of the main issues associated with Cartesian 2-D thermometry, which is hot-spot shift. The absence of this artifact, which leads to inaccurate positional temperature mapping, will ultimately allow our 3-D method to significantly reduce the time required for intra-procedural treatment planning. It will also help to eliminate uncertainty with regard to the position of the hot-spot.”
Join a global community of leaders in the radiologic sciences.
Continue your education with top-quality learning resources.
With grant applications increasing, the R&E Foundation needs you.