Get all your news in one place.
100’s of premium titles.
One app.
Start reading
The Guardian - US
The Guardian - US
Science
Richard Luscombe in Miami

AI brain map could help demystify Alzheimer’s and autism

Man with glasses on head points to screen with red and pink lines on a gray background
Dr George Spirou identifies synapses in a mouse brain at the University of South Florida. Photograph: University of South Florida

Neuroscientists at a Florida university have pioneered a technologically advanced method of brain mapping they believe can help demystify Alzheimer’s disease, autism and related disorders, and offer hope of more effective treatments for traumatic brain injuries.

A team at the University of South Florida’s (USF) auditory development and connectomics laboratory is using virtual reality (VR) and artificial intelligence to create a high-definition visual timeline of the journey of billions of neurons in the developing brains of newborn mice.

Complex imaging technology provides intricate 3D renderings of the chronology of the brain’s early formation, which are run through existing large language AI models and analyzed for changes. The rodents have similar neuron types and connections as humans.

The science is focused on the calyx of Held, the largest nerve terminal in the brains of all mammals, which processes sound. Auditory dysfunction has been widely recognized as the source of symptoms of disorders including autism that typically result in social and cognitive impairment.

“The information can help us understand serious developmental disorders that happen when the brain doesn’t develop properly early on,” said Dr George Spirou, professor of medical engineering at USF, who compared the imagery to a road map.

“It’s like you have a route from, say, New York to Chicago, and someone detours in Cleveland. You can figure out why there was some off-ramp that shouldn’t have been there, and go back and fix it.

“Maybe we’ll find the keys to some developmental disorders. And in the situation of traumatic physical injury or neural degeneration, is there a way that we can kind of recapitulate development?

“If we could trick a part of the brain into thinking it’s developing and needs to grow more synapses, that might be a therapeutic. Without succeeding totally in that realm it’s a guess, but certainly it seems reasonable.”

VR software created by Spirou, who has more than four decades’ experience in brain research, and his colleagues, is used to examine the neurons captured in the images, and analyze the synapses where they connect and communicate. Developing neural systems in mammals have been the subject of widespread study, but never at this combined level of temporal and spatial resolution, he said.

“Between the fourth and fifth gestational months, the number of neurons in the nervous system just explodes almost exponentially and synapses form at a rate of about a million per second, an incredible number when you consider there are almost 100tn synapses in an adult human brain,” he said.

“The VR platform imports huge amounts of data, and is able to look at it and understand it in 3D. There’s just no way to do it on a 2D screen.”

Spirou said that as well as possessing structural similarities with the human brain, newborn mice are used for the research because they offer a kind of microcosm of human gestation.

“At two days of age, the nerve terminal starts to grow, at four days it’s growing, and at six days of age, it’s mostly grown,” he said.

“What the brain does is like a game of musical chairs. Neurons over-innervate and then pruning takes place, like taking a chair away and someone’s out of the game. By six days of age, most of that pruning is happening, and by nine days of age it’s all set the way it will be in an adult.

“Mice are born very immature, so that first week or so in a mouse is the equivalent to time in utero in a human.”

The USF project, conducted in collaboration with scientists at University of California at San Diego, Oregon Health & Science University, and the University of North Carolina at Chapel Hill, was part-funded by a $3.3m grant from the National Institutes of Health (NIH).

In 2013, then president Barack Obama announced an ambitious human brain-mapping endeavor called the Brain Initiative (brain research through advancing innovative neurotechnologies), pledging an initial $100m in federal funds to be distributed through the NIH and National Science Foundation.

More than a decade of advances in neurological research followed, which has been mirrored outside the de facto federal umbrella. Privately funded experimentation has gained prominence in recent years and months, such as Elon Musk’s Neuralink, in which a paralyzed patient was able to control a computer by a chip implanted in his brain, before setbacks emerged.

“Other companies are doing the same thing, and even better, and studying the human brain tissue taken from neurosurgical procedures, that’s a new generation [of research], but on adults,” Spirou said.

“The timeframe that we’re looking at, which would be really four-fifths maybe into the six gestational months, we’re not there yet. It poses a whole host of issues and you wouldn’t want to take a healthy situation and perform an experiment that might alter the developmental trajectory.

“So what we’re doing with these mouse models is going to be the best approach for some time to come. What happens in science is it becomes clearer and clearer what you don’t know, and this is such a growing field.”

Sign up to read this article
Read news from 100’s of titles, curated specifically for you.
Already a member? Sign in here
Related Stories
Top stories on inkl right now
Our Picks
Fourteen days free
Download the app
One app. One membership.
100+ trusted global sources.