"Video games can become a tool for detecting Alzheimer's disease," says Antoine Coutrot.

What if a video game could contribute to the early diagnosis of Alzheimer's disease? This is the challenge of Antoine Coutrot, a CNRS researcher, linking artificial intelligence, behavioral neuroscience, and cognitive science. Since 2018, he has been working on the Sea Hero Quest project, a mobile game where the player pilots a boat through aquatic mazes.

The scientific objective: to collect, on a large scale, data on spatial navigation, a cognitive function impaired from the earliest stages of the disease. Thanks to more than four million players, the project has created a unique normative database combining navigation trajectories and demographic profiles. Antoine Coutrot is now collaborating with doctors to transform this data into a screening tool capable of identifying cognitive disorders and then referring patients to specialists if necessary.

"Currently," explains Antoine Coutrot, "to be diagnosed with Alzheimer's, the process is complex: you need an appointment at a university hospital, to see geriatricians, neurologists, to undergo neuropsychological tests, brain imaging, and sometimes a lumbar puncture. As a result, of the million people affected in France, only half are diagnosed. Video games can become an accessible screening tool outside of major hospitals: at general practitioners' offices, in nursing homes, etc. "

With over four million players, the video game Sea Hero Quest , created for Alzheimer's research, allows for a massive database, with thousands or even millions of varied profiles.

"We've known for several years now that spatial navigation, or the loss of spatial navigation, is one of the earliest markers of this disease. Alzheimer's disease is generally associated with memory loss, behavioral disorders, etc. But in reality, a few years, or even a decade or two before the appearance of these well-known symptoms, we can already begin to measure an alteration in this cognitive function that is spatial navigation."

He added: "The goal is not to make a definitive diagnosis, but to sort people according to their risk level, and to direct those who are most at risk towards the traditional diagnostic pathway. This would also help relieve pressure on specialist services."

Read also: Seven Lyon scientists awarded medals by the National Center for Scientific Research

Yes, particularly as a marker of Alzheimer's disease. We've known for several years now that spatial navigation, or the loss of spatial navigation, is one of the earliest markers of this disease. Alzheimer's disease is generally associated with memory loss, behavioral disturbances, and so on. But in reality, a few years, or even a decade or two before the onset of these well-known symptoms, we can already begin to measure an impairment in this cognitive function, which is spatial navigation.

Concretely, it is the brain structures involved in spatial navigation, such as the hippocampus for example, which are attacked by this disease.

That's it. On a journey that previously posed no problems, we start to get lost. This wasn't the case before. But it's important to point out that just because we have a poor sense of direction doesn't mean we're sick.

There are many non-pathological reasons why some people have a poor sense of direction. What is more telling is the change over time.

The big challenge of using the sense of direction in the diagnosis of this disease is that there are strong inter-individual differences that have nothing to do with the disease. They are related to age, culture, our travel habits, etc. For the doctor to be able to make a diagnosis or at least have some diagnostic aid, it is necessary to be able to compare the spatial behavior of his patient with that of a control group, composed of healthy people who are similar in demographic terms. This allows us to isolate the pathological component. For this, we need a massive database, with thousands or even millions of varied profiles, and this is…

Above all, it allows for better distribution of screening across the country. The goal is to make it possible to identify patients. Not today, as research is still ongoing, but in the coming years, this is the ambition. Currently, to be diagnosed with Alzheimer's, the process is complex: you need an appointment at a university hospital, see geriatricians, neurologists, undergo neuropsychological tests, brain imaging, and sometimes a lumbar puncture. As a result, of the million people affected in France, only half are diagnosed. The video game can become an accessible screening tool outside of major hospitals: at general practitioners, in nursing homes, etc. The goal is not to make a definitive diagnosis, but to sort people according to their risk level, and direct those who are most at risk towards the traditional diagnostic pathway. This would also help relieve congestion in specialized services.

Yes, exactly.

It's a small video game in which you play the captain of a boat, who must solve aquatic mazes. What we collect are the trajectories followed by the boat to solve these mazes. These trajectories allow us to understand what strategies are used to navigate. This gives us a window into the player's brain function, and allows us to classify those who are at risk of disease versus those for whom there are no concerns.

We are currently collecting data from patients who have been definitively diagnosed with Alzheimer's disease, but at a mild stage, that is, at the very beginning of the disease. We are comparing their behavior to that of control subjects, who are demographically similar but are not ill. The idea is to find a signature in their way of moving that allows us to distinguish them from these controls. In the long term, that is, in 10 to 15 years, the goal is for this tool to be made available to all healthcare professionals to enable early detection of Alzheimer's disease.