Autism is more complex to diagnose than many other childhood conditions. There's no physical sign or lab test; rather, making the diagnosis requires careful observation for clues such as poor language and social skills or repetitive behaviors. Standard diagnostic tests take several hours of a professional's time, and families may wait months to see someone who can assess their child.
But new research from Stanford and Harvard Medical School suggests that faster diagnoses might become possible. The research team, whose findings appear today in PLOS ONE, tested whether short home videos could be harnessed to speed the process. Using a scoring system that was pared down from the "gold standard" diagnostic test, they assessed kids' behavior in 100 short videos pulled from YouTube. About half of the videos showed children with autism; the rest did not. The scoring system classified 97 percent of the videos accurately.
The system is unlikely to replace traditional diagnostic methods, but could help relieve the diagnostic bottleneck, study author Dennis Wall, PhD, explained in our press release:
“For instance, we could use this system for clinical triage, as a way to channel traffic so that children can get the kind of attention they need as early as possible,” Wall said. Children who clearly have autism might be diagnosed primarily with videos and quickly started on therapy, freeing clinicians to spend more time evaluating children whose diagnosis is less clear-cut.
Home videos also provide information that is otherwise unavailable to those making the diagnosis, Wall said:
Another potential advantage of using video for diagnosis is that young children often behave differently in a doctor’s office than at home.
“Clinical settings are often stark, artificial and can elicit behaviors that are abnormal,” Wall said. “The odds are stacked against the diagnostic professional because the child is in an unknown environment with strangers.”
The researchers plan to explore whether the same method could also be used for making other behavior-based diagnoses, such as detecting attention-deficit hyperactivity disorder or adult-onset neurologic conditions such as Alzheimer's or Parkinson's disease.
Previously: Using Kinect cameras to automate autism diagnosis, Director of Stanford Autism Center responds to your questions on research and treatment and New imaging analysis reveals distinct features of the autistic brain