Dementias are characterised by the accumulation of various types of protein in the brain, which causes brain tissue damage and cognitive decline. In the case of Alzheimer’s disease, these proteins include beta-amyloid, which clumps together between neurons and impairs their function, and tau, which accumulates inside neurons.
Molecular and molecular changes in the brain typically start many years before symptoms appear. Dementia diagnosis can take months or even years. It usually takes two or three hospital visits and can include a variety of CT, PET, and MRI scans as well as invasive lumber punctures.
Professor Zoe Kourtzi of the University of Cambridge and The Alan Turing Institute led a team that developed machine learning tools to detect dementia in patients at an early stage. Using brain scans from Alzheimer’s patients, their machine learning algorithm learned to detect structural changes in the brain. The algorithm was able to provide a prognostic score – that is, the likelihood of the individual having Alzheimer’s disease – when combined with the results of standard memory tests.
There are currently very few drugs available to aid in the treatment of dementia. One of the reasons clinical trials frequently fail is that once a patient develops symptoms, it may be too late to make a significant difference. The ability to identify individuals at an early stage could thus aid researchers in the development of new medicines.
If the trial is a success, the algorithm could be made available to thousands of additional patients across the country.
Original Story: University of Cambridge