Can brain imaging be used to predict how brain damage will affect daily life of those living with dementia?

Lead Investigator: Dr Carole Sudre
Institution: University College London
Grant type: Junior Fellowship
Duration: 36 months
Amount: £219,527

Scientific title: An integrated neurovascular pathology score: prediction of cognitive and clinical outcome

Why did we fund this project?

Comments from members of our Research Network:

'The research looks extremely interesting and worthwhile in that it could be a very useful clinical tool in the future.'

 'I was pleased to note the researcher's direct contact with people with dementia and carers as a volunteer to balance the scientific work.'

'A worthwhile research project which has a new approach to the reason why Alzheimer's disease is generally a disease that affects older people.'

What do we already know?

The brain depends on a good blood supply to function normally. Reduction of this blood supply commonly occurs with ageing, and this means that the brain cells receive less oxygen and nutrients and are more at risk of damage. Some of this damage can be seen on special brain scans as bright spots called white matter hyperintensities. This name refers to the fact that these bright spots are found in the area below the surface of the brain where bundles of insulated 'cables' run from the surface of the brain to the spinal cord.  White matter refers to a special insulation found in some brain cells. It is important for the brain to function correctly.

In Alzheimer's disease, reduction of blood supply to affected brain areas seems to speed up the decline of thinking and memory functions. We know that damage to certain areas of the brain results in problems performing the tasks and actions controlled by that area of the brain. For example, damage to the hippocampus, the most important memory area of the brain, can result in not being able to convert short term memories into long-term ones.

Dr Sudre has already undertaken several studies to investigate changes and damage to the white matter of the brain in cases of Alzheimer's disease. Her PhD work involved designing computer programme which will be able to analyse brain scans in greater detail and improve the sensitivity of brain scans to detect white matter hyperintensities. 

What does this project involve?

Building on the work of her PhD, Dr Sudre will create a software tool that will increase the amount of brain tissue that can be analysed in one go. This will result in a brain imaging technique that can provide a prediction of how damage to brain tissue will affect the person with dementia's day to day life and impact on their future.

In order to be an effective tool to evaluate brain damage in Alzheimer's disease, the technology developed will need to satisfy three key aims. Firstly, the software must be capable of automatically detecting markers of blood vessel disease. These markers include enlarged spaces around blood vessels and scarring in the brain tissue. Automated machine learning will make the detection of the markers of blood vessel disease easier. 

Secondly, the tool must be able to provide information about the location of the white matter hyperintensities in brain cell cables. This will involve charting the location of damage in relation to the structures and architecture inside the brain. This part of the study is challenging as the method that is used to track a single brain cell cable as it moves through the brain gets lost when it hits a damage related hyperintensity. However, Dr Sudre proposes using her new method to virtually reconstruct the missing parts. This will allow her to predict which areas of the brain might be affected by the damage. 

Dr Sudre will use this information gathered by her new programme to create virtual brain software to predict how damage to a particular area of the brain would impact brain function and ability to perform tasks.

Thirdly, the programme will need to predict how damage is likely to evolve with time. This part of the study will use information from people who have had multiple brain scans over time. With this information Dr Sudre will be better able to determine whether the progression of disease and whether the disease prefers to progress along brain cell cables that are already damaged.

To meet these three aims the Dr Sudre will use three large neuroimaging datasets comprised of 8,200 people who had their brains scanned many times over a course of several years.

How will this benefit people with dementia?

This study will create a method of imaging the brain that can predict how damage to the white matter results in problems with the ability to perform day to day tasks. It will provide people living with dementia with the chance to plan for the future as they can have a clearer idea of the progression of their condition. The new software and analysis technique developed by Dr Sudre will also be useful as a measure of evaluating how well drugs are working in clinical trials.