Kāhu Takes Flight at Robinson Research Institute
This year, we had the exciting opportunity of collaborating with Daisy Hales and Ben Parkinson from Robinson Research Institute (Robinson), Victoria University of Wellington. The goal of this collaboration was to get the institute’s 1.5T Magnetic Resonance Imaging (MRI) system up and running using our Kāhu spectrometer console.
Background on the Project
Built around 15 years ago, Robinson's MRI system was originally developed as a compact orthopaedic imaging system. It was designed to showcase a new high temperature superconducting magnet design which could operate at 20 Kelvin (compared to the standard 4.2 Kelvin) and required significantly less cryogenic cooling than traditional medical systems. The system features a 1.5T magnet with a 150mm imaging volume diameter. After initial development and testing, it spent some time in Australia being used for meat and industrial imaging, before returning back to Robinson and spending two years out of use.
At the beginning of this year, Robinson teamed up with our company, Resonint, to get the 1.5T MRI system back up and running so it could take on a new role as a platform for researchers to test new MRI protocols. A core part of this project was to replace the MRI system’s existing spectrometer with our Kāhu spectrometer console which can be fully programmed with the open source Python programming language. This enables the team at Robinson to develop their own Python library of calibration tools, pulse sequences, and MRI experiments for the 1.5T system. This collaboration presented our company with a fantastic opportunity to showcase the capabilities of our hardware in a research environment and gave the team at Robinson access to hands-on MRI technology support from Resonint throughout the project.
Daisy Hales joined Robinson as a research assistant at the beginning of this year and was responsible for the bulk of the software development and imaging experiments during the project. She has an extensive background in data science, astrophysics, and biophysics with experience using Python across these fields; however this was her first time working on a project in the NMR/MRI field. The team at Resonint were really excited to have someone who was new to NMR/MRI user testing Kāhu.
My Visit to Robinson Research Institute
A few months ago, I was lucky enough to visit Robinson Research Institute and see the project for myself. When I walked into the lab, I was greeted by the humming of cryocoolers and, before long, the 1.5T MRI system joined it with its periodic chirping and clicking. The magnet was cordoned off with a yellow safety barrier and off to the left stood the giant rack of gradient drivers and amplifiers with the compact Kāhu spectrometer looking ever so small perched on top.
I was eager to see the system in action and Daisy introduced me to our test subjects - a large box of fruit and vegetables. She loaded a pineapple into the magnet’s bore and then we headed over to her laptop to see the software she had created to run the system.
It was awesome to see how much Daisy had achieved in a few months. Using Kāhu, she had created her own shimming algorithm and calibration dashboards for the system’s gradients and radiofrequency pulses. She had also written her own 1D, 2D, and 3D Spin Echo pulse sequence files and built her own dashboards to select parameters, run the sequences, and display the image data using napari (a Python based image viewer).
By varying the Spin Echo sequence's repetition and echo times, Daisy obtained proton density, T1-weighted, and T2-weighted images of the pineapple. It was really cool to see how the pineapple's fibrous core was dark in the T1-weighted image and bright in the T2-weighted images - showing that the core has longer T1 and T2 relaxation times than most of the pineapple's flesh. Daisy also imaged an orange, a kiwifruit, and a capsicum (or bell pepper). Below is a gallery showing the samples and some of the acquired MR images. You can click the image thumbnails to enlarge each image.
Fruit and Vegetable Samples
Next Steps: Eddy Currents and RARE Imaging
While the Spin Echo sequences are running smoothly, the eddy currents within the system are creating significant artifacts in the Rapid Acquisition with Relaxation Enhancement (RARE) imaging sequences. This system was built with unshielded gradients, which make the eddy currents particularly bad. The next step in the project is to measure these induced currents and implement pre-emphasis to compensate for the eddy currents’ effect on the gradient waveforms.
Make sure to keep an eye out for our next blog post where we discuss the hardware the Resonint team have designed to quantitatively measure the eddy currents and how the team at Robinson Research Institute will use this to adapt the gradient and B0 shim waveforms in the pulse sequences.
Upcoming Kāhu Resources
Once the core part of this project concludes in November, Daisy will be publishing her code and documentation from the project online. Links to these resources will be added to our website. We hope that sharing this work will benefit other Kāhu users and highlight the versatility of using the Python programming language to write pulse sequences and run MRI experiments.
This project is just another step towards empowering a worldwide growth in democratised MRI. Our goal is to continue to collaborate with researchers to build up a code and resource base for Kāhu and to create a network of idea sharing and peer to peer support. If you are using a Kāhu spectrometer console for your project, we want your input! Please reach out to us to discuss how we can support your project and how you can add your own research, code, and experience to our online resources. We can even arrange time for you to use the Robinson scanner if you want to try out Kāhu, or develop some new sequence on a 1.5 T system.
Join the Discussion!
Our forum is a great place to discuss topics with the Resonint team and other Kāhu users. If you have questions about our collaboration with Robinson Research Institute or want to share resources from a project you've been working on, send us a message in the "General" topic!
If you have questions about our product range, are interested in receiving a quote, or would like to schedule a product demonstration, send us an email at info@resonint.com.
Comments