You are here

Can a computer avatar crack the Alzheimer’s puzzle?

4 Apr 2017

LYON, 4 April 2017 -  A cure for Alzheimers? It seems far off, if not impossible. But a researcher for France’s computer science institute thinks that computer-generated avatars – not quite as glamorous as in the Hollywood movie – could at least help predict your odds of getting the disease.

Research into neurodegenerative diseases, such as Alzheimer’s, has been extremely frustrating. Clinical trials have failed repeatedly, and no drugs have been shown to even slow the progression of the disease. Such repeated failures are likely to be due in part to the strong focus on the clinical phase of the disease that starts once the symptoms of the disease are manifest. 

Silent start

But there is clear evidence that Alzheimer’s starts several decades earlier than the onset of symptoms. During this silent phase of the disease, the brain experiences a complex interplay between clusterings of neurotoxic proteins, neuronal loss, and alterations of brain structure that cause functional impairment. It is crucial to have much deeper insights into this silent phase in order to understand the disease better, find possible treatments and identify those people most at risk of developing the disease so treatments can begin earlier. 

Avatars to the rescue

Enter Stanley Durrleman, principal investigator of the LEASP project, funded by the European Research Council (ERC). He aims to develop the next generation of statistical and computational methods to build digital models of the ageing brain. These methods will combine tools from geometry, statistics and imaging science.

"Our ambition is to address this problem by the construction of digital models of the aging brain. Such digital avatars of the brain are designed to display how the brain structure, metabolism and function change during the progression of the disease,” says Durrleman.

His lab is a joint team between Inria, the national French institute for computer science and applied mathematics, and the Brain and Spine Institute (ICM), one of the main European institutes for translational neurosciences at the Pitié Salpêtrière Hospital in Paris. The LEASP project is also part of the Human Brain Project

Watch how the brain changes 

Durrleman likens the task of developing an avatar to teaching a child language. "Some people believe that the construction of such digital models, or avatars, is not possible, exactly because our understanding of the brain and disease mechanisms are so poor. However, let us take an example: we do not teach grammar and conjugation to a child who is learning how to speak. A child learns how to speak by the repetition and the combination of what he hears. It is only when the child masters the language that the child will understand its structure and its rules,” he says.

"We want to apply the same principle to construct digital brain models: algorithms will learn how the brain changes during ageing and disease progression by recombining the repeated observations of many individuals who have developed the disease in the past. The resulting model can be then interpreted to put forward new hypotheses about the disease, and can be used to make predictions about the future of the next patients."

Hope for the future

The goal of his work is to reconstruct the natural history of the disease across all stages from the observation of short periods of disease progression in several individuals.

Durrleman's research is among several studies into longevity – how long and how well we live – that have been funded over the past decade by the European Research Council, the EU’s premiere agency for frontier research. More on other ERC grantees can be found in a series of reports on longevity just launched on ERC=Science², an initiative to communicate the results of the agency’s research.

See Durrleman and other longevity researchers discussing their work live online from BioVision in Lyon, 5 April 2017, from 13 :30 to 14 :30 in the debate session, Healthy Ageing: Can Science Help You Live Longer and Better?

Watch Durrleman interviewed on Sciences et Avenir


About the European Research Council

The ERC's mission is to encourage the highest quality research in Europe through competitive funding and to support investigator-driven frontier research across all fields, on the basis of scientific excellence. The ERC expects that its grants will help to bring about new and unpredictable scientific and technological discoveries - the kind that can form the basis of new industries, markets, and broader social innovations of the future. ERC grants are awarded through open competition to projects headed by starting and established researchers, irrespective of their origins, who are working or moving to work in Europe. The sole criterion for selection is scientific excellence.

For more information go to ERC=Science² 

Media contact:

Diane Fresquez