Monkeys offer hope for autistic children

Autistic monkeys have helped local neuroscientists develop a new machine learning approach that improves diagnostic accuracy and offers fresh hope of early diagnosis.

A report by autism research center Wucailu shows that one in 100 children in China has autism. The actual number may far exceed that as the current diagnostic accuracy is only 60-70%.

Autism remains a poorly understood neurological disorder marked by impaired communication skills and repetitive behaviour patterns. It usually comes with complications, especially obsessive compulsive disorder and attention deficit hyperactivity disorder, which appear to have something in common with autism and can lead to misdiagnosis.

There are also no medical tests, such as biochemical detection or brain scans, to help doctors make a diagnosis. Doctors mainly rely on their experience. As a result, many children are diagnosed when they develop severe symptoms.

Autism is a developmental disorder. Early childhood is key period for brain growth, and early diagnosis and intervention can help children with autism get better, said researcher Wang Zheng of the Center for Excellence in Brain Science and Intelligence Technology under the Chinese Academy of Sciences.

Wang’s research team gave monkey copies of the MECP2 gene, which links to autism in humans. Better than any other lab animals, the genetically modified monkeys clearly show autism-like behaviors.

Researchers used magnetic resonance imaging to scan the brains of healthy and autistic monkeys to find out which parts are affected by MECP2. This helped researchers locate the corresponding affected areas in human brains and design a machine learning model to increase the diagnostic accuracy of autism to 82.14 percent.

The findings have been published in the American Journal of Psychiatry.