Skip to main content

Alzheimer's researchers use AI to get access to brains

The AETIONOMY project used digital technology to simulate 'personalised' brains with individual 'wiring', allowing them to study processes associated with Alzheimer's.

Video: Prof. Martin Hofmann-Apitius tells IMI about one of the biggest challenges in developing drugs for neurodegenerative diseases. 


Simulations of the brain

"How do we overcome the challenge of the inaccessibility of the human brain? Actually, in-silico-technologies help us to understand what is going on in the brain even before we can measure it directly by taking samples from living people," according to Prof. Martin Hofmann-Apitius, academic lead of the IMI project AETIONOMY.

"Based on brain imaging, it is possible to reconstruct “personalised brains", with an individual brain connectome (the 'wiring' of the brain) and other anatomical properties that are highly specific for the individual human being. We are working on this sort of simulation that even allows us to go "back in time" and to simulate the state of the brain at the age of 30 or 40, way before clinical symptoms become apparent. Modern AI technology provides us with the ability to simulate brains (not only the anatomy, but also a lot of functional aspects) and to simulate brain pathophysiology."

"’s ethically just not okay to take samples from people who have no symptoms.”

The AETIONOMY project used digital technology to study brains without needing samples from living people, allowing  them to simulate processes associated with Alzheimer's. In their mission to establish new ways of reclassifying Alzheimer’s and other neurodegenerative diseases, the team tackled the problem of how to obtain, organise, structure, integrate and interpret a broad range of distinct types of data, ranging from molecular data, to information on symptoms, available across the neurodegeneration community. They brought new structure to the classification of disease by dissecting the underlying mechanistic and molecular causes of disease, and by linking these to clinical evidence to attempt to partially validate a selection of these mechanistic drivers. 

Drug development for neurodegenerative diseases

"The reason it takes so long to develop drugs for brain disorders like Alzheimer’s is that it may take 20  to 30 years between the onset of the disease inside of cells in our brain before we start to show signs like forgetting things and being disoriented and so on. This latency between causes of the disease and the effects in a clinical context is one of the reasons it takes so long, and the other is that in neurology the human brain is hardly accessible and it’s ethically just not okay to take samples from people who have no symptoms.”

 “Humans like reductionist approaches - one cause, one reason. But there is reason to believe there are many causes of Alzheimer’s disease that may work in combination, and that was one of the biggest scientific challenges in AETIONOMY: the combinatorial causality. The project set out to answer a very fundamental question in Alzheimer’s: whether there are subgroups of patients that we can identify and that are characterised by the mechanism that gave rise to their symptoms. If you look at brains of patients who died with Alzheimer’s, you find quite diverse histopathology, highly individual paths of progression of the disease and that all indicates that there is no such thing as the Alzheimer’s disease. The task of AETIONOMY was to unravel this and I expect to see impact of our findings at least ten years down the line.”

In this video, Prof. Martin Hofmann-Apitius tells IMI about the challenges of studying the living brain. He is Head of the Department of  Bioinformatics at the Fraunhofer Institute for Algorithms and Scientific Computing SCAI, and the academic initiator of the AETIONOMY project.  

Read more

The Aetionomy project final report


Related projects: 

Manage your newsletter subscriptions
Stay informed - subscribe to our newsletter.