Introducing our paper: Author interview with Seulki Yoo

Title of release: Peer-reviewed journal publication by a GBME graduate student and alumnus.
Lead authors: Seulki Yoo (GBME, 4^th year graduate student in Master-PhD program),  Hayoung Song (GBME, Master 2019, now at Univ. of Chicago)
Title of paper: Feasibility of head-tilted brain scan to reduce susceptibility-induced signal loss in the prefrontal cortex in gradient echo-based imaging
Author interviewed: Seulki Yoo

Congratulations on your publication!

1. Please describe what you did.

We proposed a novel solution to a well-known problem in high-field brain MRI. The problem is that the static magnetic field (B0) is very inhomogeneous just above the nose/sinus cavity. That is where the bottom of the frontal brain lobe is. Traditionally, the B0 field is so non-uniform in this region that typical functional MRI has very poor quality, often with a “black hole” in part of the region. We found that if we image subjects with their head tilted backward (in a “star-gazer” or “hair-salon” position), the B0 field becomes much more uniform and much of the lost signal is recovered.

2. Please explain to the GBME students why this work is important.

The benefit of ultra-high-field brain MRI is partly negated by the presence of strong B0 non-uniformity. Prefrontal region is particularly notorious. Many researchers (tens, if not hundreds) have published methods to tackle this, by shim coils, sequence modification, and post-processing. Our method does not require any of these, but shows that simple backward head tilting can greatly improve the B0 uniformity in the prefrontal brain. We found many volunteers could easily be scanned with the new head positioning. In the future this work may enable more robust imaging in fMRI, diffusion MRI, and R2* and QSM studies in the pre-frontal region.

3. Please describe any “aha” moment, “wow” moment, and “oh, crap” moment.

- Aha! (inspiration): When we first imaged a volunteer (my advisor) in head-tilted position and saw a day-and-night difference in the EPI signal in the prefrontal region.
- Wow! (excitement): When we saw that fMRI activation was enhanced even in areas outside the prefrontal cortex, although we still don’t know exactly why. Also, when we saw that the B0 map is more homogeneous than with (simulated) 3^rd-order shimming.
- Crap! (frustration): For some volunteers the head tilting angle was limited by their chin running into the coil housing.

4. What kind of future papers should cite this?

- Any paper that discusses MRI image quality in the prefrontal cortex.
- Papers that mention various ways to improve B0 uniformity.
- Papers on multi-orientation imaging, of humans or animals.

5. To the extent you can disclose, how can this work be followed up in the future?

This work was done at 3T. I want to see if 7T MRI can also benefit from this method, even if the head coil there (Nova coil) has more limited internal space for head rotation. Also, this work may be combined with other methods to improve pre-frontal imaging, such as the one with a new spatio-temporal encoding (ERASE) from Prof. Jang-Yeon Park’s lab.

For more information, see the link to the publication:

https://doi.org/10.1016/j.neuroimage.2020.117265

Summer undergraduate research presentation

 The two-month summer vacation is almost over, and on Friday 8/21 Seok-Jin Yeo, an undergraduate junior who worked in the Lab, presented his summer research to the GBME colleagues and senior students. Mr. Yeo had successfully calculated the convolution kernels for computing the static magnetic field vectors induced by an object magnetized in an arbitrary direction, and extended gSVC -- generalized susceptibility voxel convolution -- to arbitrary orientations. The results were presented with appropriate levels of details. The students had a pizza lunch after the presentation. Good job Seok-Jin!

RF vector field mapping

 

The multi-orientation B1 mapping method successfully created a first-ever radio-frequency 3D vector magnetic field map in a liquid phantom using MRI. Hard work by Hyeong-Seop Kim to interface a custom-made surface coil with a commercial 3T scanner, unaffected by large orientation changes of the RF coil, paid off to conduct experiments reproducibly. The results are nicely visualized above in a spinning 3D GIF animation created from within Matlab.

Undergraduate research

Jisung Barg, a junior (3rd year undergrad), presented his work in the Lab in the spring semester to his colleagues on Monday 7/6. As a significant achievement, we have recently obtained a 3D vector plot of the RF field in a liquid phantom, produced by a surface coil at 123 MHz using 3T (2.895T) MRI. Mr. Barg's work was on studying the principles and visualizing the results in 3D vector arrow plots in Matlab, and process them to improve the quality of visualization. This summer there will be no intern students from outside the department, but the Lab's undergraduate research will continue with the BME students. Some new & existing ideas will be able to find their way to materialization in this way.

End of the semester, spring 2020

A highly unusual 15-week semester finally came to an end today, Friday 6/19/2020. The lab members went out for a celebratory lunch in Homaesil, and took a picture in the N center outdoor court area with masks off (!) for two minutes. Such gathering is generally discouraged, and can be frowned upon, but it was determined that any risk is sufficiently low that this modest, once-in-6-months lab activity should go head. During the semester, the BME Electromagnetics course was successfully delivered on-line, and the students' feedback has been positive. The graduate students had two papers accepted in international journals and one brand-new paper is now under consideration. The multi-orientation B1 mapping data were collected with very promising vector B1 field results, and the human 7T head coil engineering is making slow progress. Some of the research highlights will be posted soon on this site. In the summer, neither the department nor the lab has any active plans for summer internship. Exchanges and visits of intern students will generally be suppressed across the country for the year. In many labs more time will be spent for internal research and self development.

MR conferences

The two main conferences for the Korean MRI community -- icMRI and ISMRM annual meetings -- that are usually held before summer will instead take place either entirely on-line (ISMRM) or late in the year (icMRI). The official scientific conference of KSMRM, icMRI, which this year will be held jointly with the newly-formed Asian Society for MRM, was first delayed from its usual March dates to July, and then postponed again to early November. It was also reduced in length, from 3 days to 2 days, and will be held in a new location in Seoul. On the other hand, the two largest international conferences relevant to MRI, the annual meetings of ISMRM and RSNA, were announced to go fully virtual. In particular, the ISMRM meeting had been delayed from its original dates in April to August, but ultimately was decided to go on-line only. These are big changes and a great challenge for the meeting organizers. For many small academic societies, such changes are directly related to their financial well-being for the year. It remains to be seen if such on-line conferences will be more widespread from now on.

Submission results, on-line class, and KSMRM board meeting

One of the things that have always been conducted on-line is academic paper publication. Out of the four (new and revised) papers submitted in March, two were accepted without further revision (yeah!), one was rejected (sigh..), and one was given a revision request. While the team in the Lab and the collaborators worked on the manuscripts, the on-line class on electricity and magnetism went on, and already 1/2 of the course has been covered. The midterm exam, scheduled for next week, will in fact be replaced by an extended homework, to be completed remotely ("at home"), and graded with weights higher than regular homeworks.
The first KSMRM board meeting of 2020 was held on the April 13th, on-line. This is normally held during the icMRI conference at the end of March, but this year, it had to take place separately. The on-line meeting platform "zoom" was used, smoothly, to give chance for the Society's (incumbent and incoming) chairs and board members to greet each other on a video call.