Children with autism could be identified at birth… by looking into their eyes, according to new research.
It would enable treatments to begin years before symptoms develop – when they are more likely to work.
The technique uses AI (artificial intelligence) to spot abnormalities in pupil dilation and heart rate.
In tests on young girls the computer algorithm accurately detected Rett syndrome – a genetic disorder similar to autism that starts from six months of age.
And it predicted autism spectrum disorder (ASD) in experiments on mice. The US team believe it will work just as well in humans.
The device, described in Proceedings of the National Academy of Sciences, is non invasive meaning it can be used on infants without endangering their health.
Co lead author Dr Michela Fagiolini and colleagues hope the system will provide an early warning signal not just for Rett syndrome – but autism in general.
In the future, it may be used to monitor patients’ responses to treatments. Currently, a clinical trial is testing the ‘party drug’ ketamine for Rett syndrome. A gene therapy trial is also planned.
Dr Fagiolini said: “We want to have some readout of what’s going on in the brain that is quantitative, objective, and sensitive to subtle changes.
“More broadly, we are lacking biomarkers that are reflective of brain activity, easy to quantify, and not biased.
“A machine could measure a biomarker and not be affected by subjective interpretations of how a patient is doing.”
Children with ASD have problems with communicating, social interaction and are prone to repetitive behaviours.
But most cases are not confirmed until after the age of four meaning therapy is started later – delaying their potential impact.
In autism, the brain’s cholinergic circuits which are involved in arousal are disturbed triggering both spontaneous pupil dilation and constriction and speeding up the heart rate.
This happened in the mice genetically engineered to develop Rett syndrome and ASD – before they began showing any symptoms.
What is more in mice lacking a mutation in Rett syndrome known as MeCP2, restoring a normal copy in cholinergic brain circuits prevented the pupil abnormalities as well as behavioural symptoms.
Based on about 60 hours of observation of the lab rodents, the investigators “trained” a deep learning algorithm to recognise the peculiar eye patterns. It also accurately estimated cholinergic dysfunction in the autistic mice.
The team then used the formula on 35 young girls with Rett syndrome and 40 typically developing controls.
Instead of measuring the pupils, as youngsters may fidget, they used heart rate fluctuations as the measure of arousal.
The algorithm nonetheless successfully identified the girls with Rett, with an accuracy of 80 percent in the first and second year of life.
Co lead author Dr Pietro Artoni from the Boston Children’s Hospital said: “These two biomarkers fluctuate in a similar way because they are proxies of the activity of autonomic arousal. It is the so-called ‘fight or flight response.”
Autonomic arousal, a property of the brain that is strongly preserved across different species, is a robust indicator of an altered developmental trajectory, the researchers found.
Dr Fagiolini believes that together, such biomarkers could offer robust yet affordable screening tools for infants and toddlers, warning of impending neuro-developmental problems and helping to follow their progression or treatment.
She added: “If we have biomarkers that are non-invasive and easily evaluated, even a newborn baby or non-verbal patient could be monitored across multiple timepoints.”
Charities estimate around 700,000 people are on the autism spectrum in the UK – about one in every 100 people. In the US it is as high as 3.5 million. The condition is four times more common in boys.
There is no ‘cure’ but speech, language and occupational therapy are available to help children and parents, along with educational support.