A study of an iPhone app used to screen young children for signs of autism shows that the app is a novel, easy to use, and scalable method for collecting high-quality and scientifically valid data, according to a new study published June 1 in the journal npj Digital Medicine.
The app, called Autism and Beyond, was developed to address the lack of scalable, reliable, and validated tools for assessing young children’s emotions and behaviors outside of clinical settings. Built on Apple’s ResearchKit platform, the app tested a novel method for in-home screening of young children for autism. The team developed mobile technology to collect videos of young children watching movies designed to elicit autism-related behaviors, and then used computer vision algorithms to automatically code the children’s emotions and attention.
Data from this coding software, which tracks the movement of landmarks on the child’s face and quantifies emotions and attention, found that children at higher autism risk—based on caregiver reports and the Modified Checklist for Autism in Toddlers, or M-CHAT, a standard autism screening questionnaire—demonstrated a more neutral response to visual stimuli meant to prompt a positive response, such as bubbles floating across the screen or a bunny.
“The shortage of child mental health experts is a global issue, and we need new and ambitious solutions to address it,” says lead author Helen L. Egger, MD, the Arnold Simon Professor of Child and Adolescent Psychiatry and chair of the Department of Child and Adolescent Psychiatry at NYU Langone Health, who co-led the study with colleagues at Duke University Medical Center and Duke’s Pratt School of Engineering. “We seek to create apps that are feasible and scalable so that we can bring mental health resources into people’s homes and reach underserved children around the world.”
How the App Works
The entire study, from caregiver consent to video presentation and data collection, was completed within an iPhone app available for free on the Apple Store. During one year, 1,756 families with children ages 1 to 6 years old participated in the study. Caregivers completed 5,618 surveys and uploaded 4,441 videos recorded using the smartphone’s front-facing camera. Usable data were collected on 88 percent of the uploaded videos, demonstrating the feasibility of using mobile technology to collect high-quality video of children in their natural environments.
The app guided caregivers through an electronic consent process, after which they completed three to four brief questionnaires on topics such as family background, parental concerns, and potential symptoms of autism. If parents reported a high level of autism symptoms on the questionnaire, they were encouraged to seek further consultation with their healthcare providers.
Parents were then prompted, with simple pictorial directions, to show the child four short movies that have been used in earlier lab-based research with young children. The caregiver could opt to upload the whole video of their child’s responses or just the facial landmarks extracted by embedded face detection software.
Transforming Child Mental Healthcare with Digital Technology
The Centers for Disease Control and Prevention estimates that approximately 1 in 59 children has autism spectrum disorder, making it the most common childhood neurodevelopmental disorder in the United States. While mental health professionals are often able to diagnose children as young as 18 months with autism, long wait times for clinical appointments and lack of public awareness means that on average, children aren’t getting diagnosed until closer to 4 years of age.
Researchers who created Autism and Beyond sought to address this disparity by developing an app encouraging and enabling parents to identify young children’s autism risk earlier in life, allowing for earlier intervention and the ability to track changes over time. “Early detection is important because early intervention can have a significant impact on long term outcomes for individuals with autism,” says Geraldine Dawson, PhD, director of the Duke Center for Autism and Brain Development and one of the co-leaders of the study.
“We found that this app provided data consistent with what we see in a traditional clinical research setting,” says Dr. Egger. “This study lays the groundwork to create tools like this to evaluate other developmental and mental health challenges in early childhood in children’s homes.”
At NYU Langone Health, Dr. Egger has established the WonderLab, a multidisciplinary team building a digital platform for assessing common early childhood challenges including picky eating, tantrums, anxiety, and sleep in children’s homes. The WonderLab apps will provide parents with evidence-based knowledge about their young child and guided advice about how to support their child’s mental health and development and seek clinical help if needed. “Building on the success of the autism screening app, we are using technology to transform how, when, and where we give parents and all who care for children access to the best evidence-based knowledge about young children,” says Dr. Egger.
Along with Dr. Egger and Dr. Dawson, the study was led by Guillermo Sapiro of Duke and Ricky Bloomfield, now at Apple, Inc. The team included Kimberly Carpenter, Jordan Hashemi, and Steven Espinosa at Duke, high-school students, undergraduate students, graduate students, postdocs, and software developers.