Brain Program’s Mini-Workshop Called “A Landmark Example” of Academic-Industrial Collaboration


On July 23 and 24, the Focused Ultrasound Foundation’s Brain Program hosted its first invitational mini-workshop to tackle important imaging problems with the help of a world-class team of MR experts from academia and industry. This workshop was a highly successful collaborative problem-solving session that generated solutions for improving the efficiency and safety of transcranial focused ultrasound treatments.

Brain Program mini-workshop – more details

Q. What issues were discussed?

John Snell: The top three issues addressed were:

  1. MR-prescan issues: The MR console often gets ‘fooled’ by the large water bath in the brain transducer and picks very suboptimal imaging settings (the radio frequency transmitter and receiver gains). This typically results in very noisy images and/or lots of image artifacts. Because these problems were so variable and intermittent in their appearance, we didn’t know what was causing them until the mini-workshop. We learned that for now, we can work around this by training the MR techs to use the manual prescan to pick sensible gain settings during the procedure. We hope that GE can make the automatic prescan smart enough to deal with the FUS transducer and water bath at some point soon.
  2. Measuring temperature rise with the MR can currently be problematic because the thermometry images often are shifted relative to the anatomical images and the procedure for correcting this shift is very time intensive. The group discussed all the possible sources of this shift and verified that it can be reduced to negligible levels by making some changes to the way these images are acquired by the MR system (they suggested and ed several alternate pulse sequences that could be used in favor of the current one).
  3. Also the current MR temperature measurement can only be done on a single 2D plane which is presumably positioned through the center of the FUS focal spot. This creates a level of uncertainty about what the maximum temperature is (because the position of the imaging plane might not in fact precisely intersect the actual hot spot achieved by the sonication) and also limits the neurosurgeon’s view of the anatomy. The importance of monitoring temperature in three dimensions was discussed and the group ultimately went on to a pulse sequence which could acquire 5 slices every 3 seconds with equal or better image quality and significantly reduced image shift as compared to the single slice pulse sequence in current clinical use.

Although many more imaging topics were discussed, the clinical solution of even just these three has the potential of reducing the treatment time by an hour or more (currently treatment time for ET is 4-5 hours), perhaps even cutting the time in half.

Hands-on problem solving was a key part of the Brain Program mini-workshop. Shown during a work session at the UVA Focused Ultrasound Center are: (standing left to right) Dennis Parker (Utah), Chang-Sheng Mei (Harvard) and Craig Meyer (UVA); (seated, left to right) Kim Butts-Pauley (Stanford), Anthony Vu (GE Healthcare), Matt Eames (FUS Foundation) and Eyal Zadicario (InSightec).

“Some of the solutions identified during the workshop will be applied in the very next brain patient procedure,” says John Snell, PhD, Brain Program technical director. “The solutions are expected to reduce treatment time by improving the quality of the MRI scans.”

Conducted at the University of Virginia’s Darden School of Business, the workshop brought together representatives from industry (InSightec and GE Healthcare) and neurosurgeons, MR engineers and MR physicists from leading academic centers in the U.S. (Harvard, Stanford, Utah, the University of California San Francisco, UVA and Vanderbilt) and Europe (Institut Langevin and University Children’s Hospital, Zurich).

“The workshop was highly productive in finding concrete solutions for many technical issues and serves as a great example of how the Foundation opens up channels of communication and catalyzes collaboration among academic researchers, clinicians and industry. We hope to replicate this experience in the future.”

– John Snell, PhD
Brain Program Technical Director
Focused Ultrasound Foundation

Following the workshop, Eyal Zadicario, director of Neuro Programs for InSightec, wrote, “Getting this team together is a great achievement for the Foundation and a landmark example of collaboration between academia and industry.”

Workshop focused on real cinical needs

The workshop kicked-off with a clinical update from W. Jeffrey Elias, MD, the neurosurgeon who led the recent essential tremor study at UVA. He discussed the evolution of treatments for Parkinson’s disease and essential tremor, presented the essential tremor study results, described step-by-step how procedures were performed and indicated where imaging improvements could significantly decrease treatment time and increase safety. His presentation served to focus the subsequent workshop discussions on real clinical needs.

Next, Zadicario provided an overview of the technical aspects of the ExAblate Neuro system and presented a set of case reports illustrating the imaging issues experienced at all brain sites.

“The first day saw a lively, collegial discussion among all participants. The tone of the meeting was completely different from what we typically see at a presentation-driven meeting,” Snell observes. “It was great to see all the participants working together as a unified team to solve the presented problems.”

Attendees spent the second day of the workshop at the UVA Focused ultrasound Center engaged in hands-on experiments and problem-solving. “They determined that a really meaningful number of imaging artifacts/issues can be addressed in the short term. For example, the group demystified sources of several signal-to-noise and image artifact problems and went on to solve them during the following hands-on session. This will immediately improve image quality in the very next patient treatment,” Snell reports. “Other imaging solutions and improvements proposed during the workshop will require vendor implementation, but clinical availability shouldn’t be far off.”

Dialog is continuing via Foundation’s online Collaborative Research Network (CRN)

While face-to-face discussions have ended, workshop participants are continuing their dialog via the Foundation’s online CRN. In addition to posting a digest of workshop discussions and the MR data acquired during the meeting, they continue to exchange ideas and suggestions online. As Snell observes, “The CRN provided real value during and following the meeting by documenting the hands-on discussion and results and by fostering ongoing dialog about the issues we addressed. It has also been highly useful as a tool for disseminating the results of the meeting.”

Written by Ellen C., McKenna

Q. What longer-term research projects were identified?

John Snell: Three projects were identified:

  1. Volumetric thermometry – The project goal would be fully three-dimensional acquisition of MR thermometry data, perhaps even over the entire skull/brain.
  2. Acoustic radiation force imaging (MR-ARFI) – This is a way for the MR to make images of the sonication focal spot without heating. Applications include spot verification for future treatments that don’t involve heating like blood-brain-barrier opening and clot lysis for stroke. This technique may also one day allow us to automatically correct acoustic aberrations caused by the skull without the need for CT. We currently have to use a pre-operative CT scan to calculate these corrections so that the skull doesn’t ‘blur’ the acoustic focus of the transducer. During their hands-on work at the University of Virginia Focused Ultrasound Center, mini-workshop participants were able to install and a research ARFI pulse sequence that was created by the Stanford University team. We were able to use this pulse sequence to create an MR image which clearly showed the sonication focal spot in a gel phantom mounted inside an ex-vivo skull, with no heating.
  3. Development of a dedicated FUS-compatible head coil – Completing the first two projects on this list will ultimately require a dedicated FUS-compatible head coil. Such a coil would actually benefit all MR imaging with FUS. By basically placing a radio antenna close to the head, we can collect a lot more signal than we are currently capable of doing with the antenna built into the bore of the MR, the so called body coil. The increase in signal will allow us to someday either create images with significantly increased quality or to create images faster. In the near term, images with less noise and more visual detail and contrast would be welcomed by neurosurgeons. Faster data acquisition will ultimately (we hope) allow us to do things like volumetric thermometry fast enough to monitor temperature rise in real time. Several groups are currently developing and ing FUS-compatible head coils. We hope the discussions during the mini-workshop aroused enough interest with GE and InSightec to have them ultimately produce a clinically usable head coil.