Computers can now spot autism in people using facial cues

A newly-developed computer algorithm that analyzes facial expressions as well as head movements has recently been developed, which could possibly aid doctors with diagnosing conditions like attention deficit hyperactivity disorder and autism.

When it comes to testing for such disorders, experts usually employ standard questionnaires, and look for certain behavioral markers using manual observation of the patient. “These are frequently co-occurring conditions and the visual behaviours that come with them are similar,” according to Michel Valstar, one of the research's proponents at the University of Nottingham, UK.

To make things easier, Valstar and his team utilized machine learning to be able to better observe behavior among people. They gathered 55 adults, who read and listened to stories, while also answering a few questions about themselves. The proponents also monitored head movement to know how much the subjects' attention wandered. After the observation, the subjects were classified into four categories: those diagnosed with autism-like conditions, ADHD, both, or neither.

Through this collected data fed to the computer, they learned how to differentiate the responses of respective groups. People with both autism-like disorder and ADHD were found to be less likely to raise their eyebrows when they come across surprising news. The computer with this information was able to correctly identify those with autism-like conditions or ADHD with 96% accuracy.

Eric Taylor of King's College London affirms that the study can indeed aid as a diagnostic tool when it comes to these disorders. However, according to him, observing kids in everyday life is still the best way to observe for possible conditions of ADHD or autism.

"Algorithms won’t take over from doctors any time soon," says Valstar. “We are creating diagnostic tools that will speed up the diagnosis in an existing practice, but we do not believe we can remove humans. Humans add ethics and moral values to the process", he concluded.

Back in 2014, in an effort to aid kids who are suffering autism, Samsung developed an app that helps them focus, concentrate, and most importantly, practice eye contact with others.

Source: Cornell University Library via New Scientist | Image via Autism Society of North Carolina

Report a problem with article
1483614709_nvidia_spot
Next Article

Nvidia announces new Shield TV and companion Spot peripheral with Google Assistant support

1483608555_dsc01468
Previous Article

'Endless' debuts new PCs for the US with $129 Mission Mini and $249 Mission One

4 Comments - Add comment

Advertisement