Noises made by babies can be used to identify whether they have autism, according to scientists in the US.
Those with the disorder apparently babble different sounds than those without and, using voice analysis technology, researchers say they can single out differences with an accuracy rate of 86%.
The same method can also identify infant children whose language development is impaired, the scientists say.
Autism is defined as a spectrum of lifelong conditions relating to a person's development, which can be characterised by any or all of the following: an inability to relate to or communicate with other people, having little or no social skills, repetitive behaviour, obsessional traits. Around 500,000 people are said to have some form of the disorder in the UK.
Autism expert, Professor Steven Warren at the University of Kansas, who was involved in the research, said: "This technology could help paediatricians screen children for ASD (autism spectrum disorder) to determine if a referral to a specialist for a full diagnosis is required and get those children into earlier and more effective treatments."
The study looked at around 1,500 vocal tracks recorded all day by devices stuck to the clothes of 232 children aged between 10 months and four years.
The Proceedings of the National Academy of Sciences journal published the study findings.