Eye-Tracking Device Predicts Expert Autism Diagnoses

Warren Jones, PhD

An eye-tracking diagnostic tool provides measures of social visual engagement that which specialists may use to affirm autism diagnoses in suspected toddlers, according to new research.¹

Data from a team of US investigators show a device from EarliTec Diagnostics—recently authorized by the US Food and Drug Administration (FDA) for detecting autism spectrum disorder²—was associated with consistent specificity and sensitivity. The findings implicate an opportunity to shorten the average time to autism diagnosis in the youngest identified patient population, which in part would improve the behavioral treatment opportunities for children and assuage diagnosing specialists who have been historically inundated with referrals.

Investigators led by Warren Jones, PhD, of the Marcus Autism Center at Children’s Healthcare of Atlanta, sought to evaluate the performance of eye-tracking measurements focused on social visual engagement, relative to clinical diagnoses by specialists in young children referred to autism clinics.

In an interview with HCPLive’s podcast series DocTalk, Jones explained the reliance on behavioral traits in modern autism diagnoses. Best practice guidelines call for an assessment of verbal and cognitive abilities along with detection of social disability—all of which are highly validated means of interpreting autism.

“However, there’s a big challenge in the community that there are not enough expert clinicians to meet the needs of all the children who have concerns about their early development,” Jones explained. “So there are really long waitlists at almost all of the major autism treatment centers and specialty diagnostic centers across the country.”

Over the last 2 decades, an increasing awareness of autism’s prevalence based on behavioral and social traits has significantly increased the rate of parents and guardians seeking an early diagnosis. Despite this he median age of diagnosis has not budged from the approximate 4-year range—a “direct contrast” of the average age of 2 years at which parents first become concerned about their child’s behavioral development, Jones said.

The concept of eye-tracking tools was informed by clinical insights into the trend of high-risk children showing fundamental social disability and reduced attention to social cues, Jones explained. His team’s intention is not to replace expert clinicians’ assessments, informed by thousands of pediatric consultations—but rather to best quantify those qualitative insights.

“In many respects, what we’re doing is not unlike a treadmill test in cardiology, where in order to identify some challenges, a patient might run on a treadmill for a little while—elevate the heart rate—and at that moment you can identify some differences that maybe aren’t identifiable in other straits,” Jones said.

Their analysis included children aged 16 - 30 months old enrolled at 6 specialty centers across the US from April 2018 - May 2019. Care staff were blinded to clinical diagnoses of autism spectrum disorder and required to use the automated EarliTec device to measure eye-tracked-based visual engagement of children who were prompted with social cues, such as the sound of their own name or children playing together.

Trial participants additionally received expert clinical diagnoses based on best-practice standardized protocols from the blinded care team.

Jones and colleagues successfully tracked eye movement for social visual engagement in 475 children. Mean age was 24.1 months old and three-fourths (74.1%) of children were Black. In all, 221 (46.5%) participants were diagnosed with autism.

Investigators observed a social visual engagement sensitivity of 71.0% (95% CI, 64.7 - 76.6) and specificity of 80.7% (95% CI, 75.4 - 85.1) among all children assessed with the eye-tracking device. Among the children with an autism, the device showed 78.0% sensitivity (95% CI, 70.7 - 83.9) and 85.4% specificity (95% CI, 79.5 - 89.8).

The team confirmed the eye-tracking test results correlated with the clinical assessments for individual levels of social disability, as well verbal ability and nonverbal cognitive ability. They additionally noted the “diagnostically complex sample” of children included in the analysis, of which the spectrum of autism was very broad.

“The challenge of reference standard differential diagnosis in this sample was highlighted by clinician uncertainty in 29.5% of children, which is comparable with, yet slightly lower than, other reports in the literature (of up to 40%),” investigators wrote. “The present results suggest that measurement of social visual engagement—how children look at and learn from the surrounding social environment—may offer such a biomarker for autism.”

Though the trial was limited by the lack of intent to evaluate general population, the team concluded the eye-tracking device was predictive of autism diagnoses by clinical experts.

“Further evaluation of this test’s role in early diagnosis and assessment of autism in routine specialty clinic practice is warranted,” they wrote.

References

1. ^{Jones W, Klaiman C, Richardson S, et al. Eye-Tracking–Based Measurement of Social Visual Engagement Compared With Expert Clinical Diagnosis of Autism. JAMA. 2023;330(9):854–865. doi:10.1001/jama.2023.13295}
2. ^{EarliTec Dx. EarliTec Diagnostics Receives FDA 510(k) Authorization for Next-Generation EarliPoint Evaluation for Autism Spectrum Disorder. Press release. Published July 13, 2023. https://www.earlitecdx.com/2023/07/13/earlitec-diagnostics-receives-fda-510k-authorization-for-next-generation-earlipoint-evaluation-for-autism-spectrum-disorder/}