Google Glass with machine learning helps autistic children read facial expressions
Stanford Medicine News Center

Google Glass with machine learning helps autistic children read facial expressions

Although a commercial failure, Google Glass lives on as a powerful medical research tool

Like GearBrain on Facebook

It may have failed to win over consumers, but Google Glass is enjoying a second life as a powerful medical research tool.

The latest application for Google's augmented reality wearable is to help autistic children better-understand the facial expressions of their friends and family.

Read More:

A pilot study run by Stanford University saw 14 families with autistic children try out Google Glass and an Android smartphone, each running a custom-built system called Superpower Glass.

The app uses the Glass' outward-facing camera to focus on the facial expressions of people the child interacts with. Machine learning is then used to understand what these expressions are, before the child is told - via the wearable device's speaker and display - what the expression means.

There are eight core facial expressions which the app deals with. These are happiness, sadness, anger, disgust, surprise, fear, neutral and contempt - the latter is described in more child-friendly language as 'meh'.

Children normally pick up these facial expressions and their meanings by engaging with the people around them, but for those with autism it's more difficult as they fail to pick up on the emotions of others without focused treatment.

Dennis Wall, PhD, associate professor or pediatrics and of biomedical data science, said in a report by Stanford: "We have too few autism practitioners. The only way to break through the problem is to create reliable, home-based treatment systems. It's a really important unmet need."

A commercial failure, Google Glass is enjoying a second life as a medial research tooliStock

The app has two modes. In 'free play', children wear the Google Glass while interacting with their families, and the software provides the child with a visual or audible cue each time it recognizes one of the eight emotions on the faces of people in view of the camera.

The second mode introduces a gaming element by asking the child to guess the facial expression shown by a parent, who is asked to act out some of the eight emotions the Glass can understand.

The Stanford Medicine report explains: "The game helps families and researchers track children's improvement at identifying emotions. In 'capture the smile,' children give another person clues about the emotion they want to elicit, until the other person acts it out, which helps researchers gauge the children's understanding of different emotions."

The app helps explain expressions with words and emojiStanford Medicine

The results of the trial have been positive. Not only did the families find the system engaging, useful and fun, but children were willing to wear the Google Glass, and the device stood up to the wear and tear of being used by youngsters.

Most interestingly of all, the pilot - which was tested on children aged between three and 17 - saw an average reduction of 7.4 points on the autism spectrum. Six of the children recorded scores so much lower than before, that their autism was moved to a less severe class. Four had their autism downgraded from severe to moderate, while one was reduced from moderate to mild and another moved from mild to normal.

Wall added: "Parents said things like 'a switch has been flipped; my child is looking at me.' Or 'suddenly the teacher is telling me that my child is engaging in the classroom.' It was really heartwarming and super-encouraging for us to hear."

Like GearBrain on Facebook
The Conversation (0)

GearBrain Compatibility Find Engine

A pioneering recommendation platform where you can research, discover, buy, and learn how to connect and optimize smart devices.

Join our community! Ask and answer questions about smart devices and save yours in My Gear.

Top Stories

Weekly Deals