Predicting dementia risk using smart software

Currently, the program developed by researchers Matthijs Biesbroek and Mirthe Coenen can only be used in patients who already have cognitive complaints, such as the onset of dementia. However, in the future, it should also be possible, the researchers hope, to use this intelligent diagnosis in people who do not yet have any complaints. For this, the intelligent program must be developed, in which the use of AI (artificial intelligence) is also being considered.

The effects of aging and cognitive problems

The white matter of the brain is the “fiber optic network” through which all information flows. The substance consists of axons of nerve cells, surrounded by a white, fatty substance, myelin, that protects the axons as a kind of insulating material. Aging is one of the important “enemies” of white matter. As a result, tissue is damaged and cognitive problems can arise, among other things.

The consequences of these injuries can be different for each patient. The consequences experienced by the patient depend particularly on the site of the damage. Sometimes people have a lot of white matter damage, and they don’t suffer from it. Other times, people function less well than before, while the damage to their white matter is relatively minimal,” explains neurologist and researcher Mattis Bisbroek.

Predict the risk of dementia

Thanks for that Research And the UMCU team’s smart software, it is now possible to quantify and predict the consequences that white matter could have on individual patients. The researchers first succeeded in identifying the major nodes of the white matter.

They then used this information to develop intelligent software that could analyze brain scans. Thus, this software is able to predict complaints, such as dementia, that patients may develop in the future due to white matter damage.

“To determine which of these white matter pathways are important and which are less important, we needed a large cohort of patients,” Mattis says. For these patients, both a brain scan and cognitive tests had to be provided. In total, the Utrecht researchers used data from 3,525 patients from Europe, Asia and the United States.

Further development using artificial intelligence

The software as now developed can also be used to analyze standard brain scans such as those performed at a memory clinic. But, as already mentioned, the program is only suitable for analyzing people who actually have problems.

Researchers are already working to further develop the program. In addition, it should be possible to go through all the steps of the process completely automatically. They even have plans to apply artificial intelligence (artificial intelligence) to give the software the ability to learn. The program is expected to be ready in early 2023.

Using AI applications and algorithms to improve and speed up the diagnosis of brain problems, such as dementia, is nothing new. The first steps in this field were already taken years ago. One of the most recent initiatives is the International Initiative artificial intelligence project. She wants to develop smart, AI-assisted digital tools to examine brain connections and estimate dementia risk in people with mild cognitive problems.