Tracking dementia in brain networks
While just at the beginning of her career, Anne Hafkemeijer can already be called an all-round LIBC member. Hafkemeijer received her PhD in May 2016, as part of the research group of LIBC director Serge Rombouts, with a dissertation on methods for detecting dementia in brain networks. After her PhD defence, she stayed on at LIBC as part of the LIBC support team, where she assists researchers from different faculties with their MRI research.
‘I almost prefer my work as MRI research coordinator to conducting my own research,’ says Hafkemeijer. ‘You can really make a difference this way and speed up the research process’.
‘We help to improve MRI research’
The Leiden Institute for Brain and Cognition has its own MRI scanner, located in the LUMC. The LIBC support team has a central role within the institute.
‘When a researcher sends us a research proposal, we check all safety matters involved, schedule the scan time and analyse the MRI data,‘ says Hafkemeijer. ‘Most researchers are familiar with MRI research and conduct their own data collection and analysis. But for others less familiar with MRI research, we can help improve their research.’
Researchers can become shareholders of the LIBC scanner. At the beginning of the year they decide how much scan time they will need. Half of the applications for scan time come from the Faculty of Social Sciences. The other half is shared by different disciplines such as linguistics, radiology and psychiatry. According to Hafkemeijer: ‘It is our task to connect these people’.
The demand for the LIBC scanner keeps growing. ‘If you offer these services, you automatically attract more research,’ says Hafkemeijer. In the summer of 2017 the LIBC scanner will be almost completely renovated and upgraded. Only the heavy magnet, the heart of the scanner, will remain the same.
Dementia’s fingerprints on the brain
Anne Hafkemeijer herself has had her share of scan experience. In Rombouts’s research group, her dissertation work focused on MRI methods for tracking the progression of different kinds of dementia in the brain. She compared Alzheimer’s, the most common kind of dementia, and frontotemporal dementia, a disease known for causing changes in personality and behaviour. Anne also studied the effects of normal aging on brain networks.
‘In late stages, the symptoms of Alzheimer’s and frontotemporal dementia clearly differ,’ says Hafkemeijer. ‘But in early stages of the disease, when one has just been diagnosed, symptoms can be ambiguous and tend to overlap, causing difficulties in simple, day-to-day tasks like cooking.’
She therefore focused on one of the burning questions in brain and aging research: Is it possible to distinguish different kinds of early stage dementia in the brain?
To tackle this issue, she focused on existing brain networks. In recent years, it has been found that cognitive functions like language, memory and behaviour are not only localised in specific brain areas, but also entailed in larger so-called brain networks.
Hafkemeijer followed patients over the course of the disease and studied functional as well as structural networks. ‘This approach is fairly new. We were one of the first research groups in the world to use a new technique that allows you to observe structural networks.’
Research with an emotional impact
The results were promising. Differences between Alzheimer’s patients and patients with frontotemporal dementia were clearly seen and were even more evident in structural networks than in functional networks. ‘The difference between the patterns that effect the brain were clear.’
Healthy participants also showed damage in specific networks caused by aging. ‘For example, the subcortical areas in the middle of the brain, the areas for emotions and memory, were affected. You can see decreased connections within the network.’
There was another group of interest: people visiting the memory clinic because they noticed that their memory is deteriorating. ‘Cognitive tests do not yet show alarming results, and they are not diagnosed with dementia. But on MRI scans we were able to detect a difference, in the structure of specific areas as well as in functional networks. It would be interesting to have these people come back in a few years to see whether these findings were really an indication of early stage dementia.
This research also had an emotional impact. Hafkemeijer saw the patients’ condition worsen over the course of her research. ‘At first we scheduled the follow-up scan two years after the first scan, but some patients would not make it two years, so we had to adjust our follow-up session to one year after the baseline scan.’
Using brain scans to predict the chance of developing dementia
The ultimate goal is to be able to diagnose every individual patient as early as possible. Although dementia is not curable, it is important for patients, their family, friends and healthcare workers to learn how to deal with this disease in an early stage. Above all, medication to slow down the disease is most effective in the early stages when there is only limited damage.
Hafkemeijer: ‘We’re not there yet; people’s brains differ too much. For now we have to work with group averages.’ Serge Rombouts’s research group is now studying the possibility of teaching a computer system to measure the chances of developing Alzheimer’s disease, based on brain scans.