We are happy to share a new lay summary of a recent STARSTEM publication, Magnetic resonance fingerprinting of the pancreas at 1.5 T and 3.0 T.
Dr Joshua Kaggie and team members from the University of Cambridge recently had this journal article published in Scientific Reports. Scientific Reports is an open access journal publishing original research from across all areas of the natural sciences, medicine and engineering.
We asked the team to discuss their work.
What were you aiming to find out in this publication?
We would like to acquire MRI data faster and with higher quantitative accuracy. This work uses ‘MR Fingerprinting’, which combines MRI physics simulations with pseudorandom acquisitions to accelerate MRI.
Why is this important?
When MRI is performed, it is normally done in a non-quantitative manner using methods that have been relatively unchanged over the past few decades. MRI is used to diagnose an increasing number of diseases, which requires faster methods that are also precise. MR fingerprinting is a method that helps address this challenge by accelerating MRI throughput, while at the same time providing quantitative data.
In terms of iron-oxide imaging, this method will help improve the quantitation.
Describe the methods chosen.
In this work, we image the pancreas with MR fingerprinting and a challenging area to image due to its central location in the body – deep and far from the receivers placed outside the body.
We imaged 16 normal volunteers with MR fingerprinting. MR fingerprinting uses pseudorandom acquisitions that are combined with MRI physics simulations to accelerate the imaging. The images were acquired at two MRI field strengths common in hospitals, 1.5 T and 3.0 T, in order to demonstrate the feasibility on typical systems. The physics simulations required several hours, after which they could be applied to the data from each subject. The images were acquired in 2-4 minutes, which is a rapid time for quantitative parameter mapping.

These quantitative MRI maps of the human pancreas were acquired with MR fingerprinting. T1 and T2 are measurements of MRI signal decay, which are altered in diseased/healthy tissues, or in the presence of iron – such as with iron-oxides used in the STARSTEM project.
How could this work benefit patients?
The intended impact is to improve patient throughput for routine MRI scans, while at the same time acquiring higher quality data. When successful, this will enable faster routine imaging, which will translate to cost savings, as well as a better diagnosis by improving the images (or maps) that are acquired with MRI. This also feeds into machine learning techniques, which benefit from higher repeatability between centres.
The main output of this work is demonstrating that we can reliably obtain MRF maps within the abdomen, which is a challenging area to image. This area is subject to breathing motion, but also small variations in magnetic fields that have large effects on images – including air in the lungs and bowels.
In addition, this work showed that these images could be obtained while a patient is free-breathing. Breathing artefacts are common to MRI in the abdomen, and having a technique that can ignore breathing is powerful, particularly when we want to image animals that can’t hold their breath on command or patients who have difficulties holding their breath for long periods.
Who were your collaborators?
This work was done in (external) collaboration with GE Healthcare and the University of Pisa.
What are the next steps?
The next step is to apply MR fingerprinting in sheep stifles (=knees) to measure the changes induced with iron labelled stem cells. The additional measurements with this are the presence of iron in abdominal sheep organs.
You can read the full paper online or download a PDF of the paper on our publications page.