Can It Be Done? Use of Facial Recognition Software in Virtual Platforms such as Google Meet for Children with Autism

Shawn Yehudian, MD

Cambridge Health Alliance
Can It Be Done? Use of Facial Recognition Software in Virtual Platforms such as Google Meet for Children with Autism

Scientific Abstract

Background: Social barriers for children on the autistic spectrum include deficits in identifying social cues and metalization; Play and psychodynamic psychotherapies help grow this capacity. Some studies suggest that children on the autism spectrum may have an easier time identifying facial expressions on caricatures versus human expressions. There may be an opportunity to utilize this ever-advancing technology to help children on the autism spectrum more accurately be cued in to facial expressions of their therapists. Can we adopt and integrate this technology into our video telehealth platforms in order to meet the needs of our population of autistic children during (and after) the COVID-19 pandemic?

Methods: A literature review was conducted on Hollis using three separate literature searches with different combinations of the following keywords: autism, facial expression, emoji, emotion, wearable, smart glasses, understanding, and “computer-based intervention” from 1998 to March 2021. This search resulted in a total of 408 articles. Non-peer-reviewed articles were excluded. Inclusion criteria included peer-reviewed articles (n = 184). Thirty-Nine of those articles were relevant to the topic (n=39).

Results: There have been studies on the use of technology to analyze emotional facial expressions and translate those emotions in the form of visual emojis or avatars for children on the autistic spectrum. One 2015 study on a software that utilizes face processing instruction concluded that by practicing eye gaze, joint attention, emotional cognition, and facial recognition skills on a computer, participants were able to improve their ability to recognize basic emotions such as happiness, sadness, neutrality, anger, disgust and fear, and also enhance social skills and mentilization abilities.

Conclusions: Adoption of facial recognition software to video telehealth platforms, if achievable, may be instrumental in helping children on the spectrum. By providing virtual emojis that match facial expressions in real time, children on the spectrum may have the opportunity to connect with their therapists on a deep level, connecting emotions with expressions.

Live Zoom Session – April 21st

research Areas


Shawn Yehudian, MD, Maril Yehudian, MD