Pawel Markiewicz is a Postdoctoral Research Associate based at University College London. He is currently working on a Dementias Platform UK study that measures scanning variability at and between each of the sites in the DPUK Imaging Network.
“Each of the seven participating DPUK sites is equipped with state of-the-art PET-MR scanners, allowing scientists to picture the brain in two vital and complementary ways: using Positron Emission Tomography (PET) and Magnetic Resonance (MR) simultaneously. Doing these different types of scans together is new and exciting for researchers but important checks need to be done – and that’s where the phantoms come in.
“PET phantoms are designed objects, usually in the shape of a cylinder, which are filled with a water solution of a radioactive tracer. I use phantoms to evaluate different scanner properties to help establish the ground truth for each of the scanners. By comparing what I know about the properties of the phantom with the image produced, I can evaluate the scanner’s performance. I use different phantoms for different tests. A cylindrical one helps me test the uniformity across images; the one with inserts of different sizes helps me understand how the scanner will pick up contrasts in different areas of the brain.
“Characterising the scanners like this is particularly important for interpreting differences in human brain scans – for example the distribution of amyloid-beta – a protein that is known to be involved in the development of Alzheimer’s disease. The key aspect of phantoms is that we do know the underlying true distribution of the radiotracer, whereas for real brain scans we do not know it. When we conduct the PET-MR scan it’s critical to know that our measurements are accurate and there is no interference or quantification errors.
"I’ve now completed my phantom work in all the sites, all the way from Edinburgh to London. Now we are ready for the next step, which is a pilot study doing repeat scans on healthy volunteers. The aim is to harmonise and quantify the measurement variability within each site and across sites with similar and different scanning equipment. We take a lot of care to ensure the network is calibrated and the result is that we can be sure, to the best of our ability, that scientists in research institutions across the UK are able to work with image data of the highest quality and reliability. With these scanners we will be able to see the precise character of amyloid build up, for example, and exactly where it is. It’s in this level of detail that new insights into treatment will come.”
What to read next
Although the more commonly known ‘hackathon’ has been used to address societal-level problems from homelessness to corruption, we think that it’s only recently that the data science community is turning its attention to dementia. The DPUK datathon series might well be one of the first for the field in fact. Chris, DPUK’s Data Project Manager, and Sarah, DPUK’s Senior Data Manager, take us through the three key ingredients for a successful datathon.