Eight models developed by researchers in the Americas have been examined by automatic learning. By testing autism on home videos, experimentation is quicker and more efficient. "According to a study published by scientist POS Medicin.
Each model contains "a set of algorithms which includes the qualities of children 5 to 12. A general score indicated whether the child is autism".
Considering the evaluation of the models, the children's faces and hands were shown and their "social interaction and use" asked families who had been recruited for study to send a five-minute home videos of toys, pencils and bowls.
From those pictures, 116 children were diagnosed with autism at an average age of 4 years and 10 months. It was developed two years (average two years and 11 months).
An expert expert analyzed videos with "query" or "no" answers using a query of 30 questions based on the typical autism nature combined with eight mathematical models.
The best results are 94.5% of children with sample autism and 77.4% of non-oriental children.
For a verification of the results, they evaluated another 66 videos.
The same model has a child with 87.8% autism and 72.7% of the disease.
"Across the United States, the average waiting list to get access to a standard service (for childhood Atrisz) can be up to a year," said the Wall.
Another advantage of using home videos for diagnosis is that they "bring the child to their natural environment". In a medium different from that of a "consistent, fictitious and contradictory" medium.
"We've proven that we can identify small groups with differential qualities." In professional consultations, experts can quickly and independently evaluate within a virtual online environment within minutes. "