Camera on Google Glass May Boost Social Skills in Kids With ASD

Norra MacReady

March 26, 2019

A learning aid that can be incorporated into Google Glass and accessed with the help of a smartphone app is showing promise in the treatment of children with autism spectrum disorder (ASD).

In a small, randomized clinical trial, children who wore and used the device for 6 weeks in addition to receiving usual treatment scored higher on at least one socialization subscale compared to children who received only usual treatment, lead author Catalin Voss, MS, and colleagues report in a study published online March 25 in JAMA Pediatrics.

"Our hope is that this can show us how to create solutions through the use of more ubiquitous technologies for families who are waiting for care," senior author Dennis P. Wall, PhD, told Medscape Medical News.

Currently, he explained, families often face long wait times for one-on-one services from interventionists who specialize in applied behavior analysis (ABA), the current standard of care for children with ASD. Such care can also be very expensive.

"This trial shows us that digital therapies can work in this patient population," he said in an interview.

The device, called Superpower Glass (SG), consists of a small camera attached to the Google Glass frame. It is described as an "artificial intelligence–driven wearable behavioral intervention." The camera is programmed to recognize faces and facial expressions and to give the wearer a voice audio cue when it detects one of eight emotions: happy, sad, angry, scared, surprised, disgusted, "meh," and neutral.

It is synchronized to an Android app that coordinates face tracking and emotion detection. Families can disable the vocal cues and rely on emotions that appear within the child's visual field to alert him or her to the appropriate emotion.

Future versions of the software may be used with iOS devices and wearables other than Google Glass, said Wall, who is associate professor in the Departments of Pediatrics, Psychiatry (by courtesy), and Biomedical Data at Stanford University, California.

The app can launch games that encourage children to elicit a particular emotion in a family member's face (for example, by telling a joke to elicit a happy smile), to guess an emotion a caregiver may be acting out, or to engage in unstructured play in which the child receives cues from multiple people. Children and their families or caregivers can choose which of the games they wish to play at any given time.

Children in the treatment group were asked to try each of these engagement activities at least once. The recommended treatment schedule was 20 minutes three times per week with family members, plus once a week with the ABA therapist, for a total of four sessions per week.

The study is "very cool — small but well designed, and the app is very nicely done," Susan E. Levy, MD, professor of pediatrics, the University of Pennsylvania, and the Center for Autism Research, Children's Hospital of Philadelphia, said in an interview with Medscape Medical News.

"To me, it's encouraging that this type of intervention, which uses a form of virtual coaching, may be appealing to some kids with autism," said Levy, who was not involved in this research.

However, said Levy, "I don't know if the differences they observed will persist over the longer term. They may need a larger study to examine that."

Low Participation Adherence to Recommended Use

In their study, Voss, of the Department of Computer Science at Stanford, and coauthors randomly allocated children with ASD who were 6 to 12 years of age either to use the SG device along with their usual ABA treatment for 6 weeks or to continue with ABA by itself. To be eligible for the study, the participants had to be have been receiving ABA therapy at least twice a week and to have achieved a score higher than 15 on the Social Communication Questionnaire.

The primary outcome measures were changes in scores on three standard tests:

  • The Social Responsiveness Scale, 2nd edition (SRS-II), which identifies the presence and severity of social impairment across five domains;

  • The Vineland Adaptive Behavioral Scales, 2nd edition (VABS-II), socialization subscale, which measures communication, socialization, motor function, daily living, and adaptive behavior skills and is often used to track changes over time;

  • The Developmental Neuropsychological Assessment, 2nd edition (NEPSY-II), affect recognition domain, which measures the patient's ability to recognize emotions in facial expressions from a standardized set of photographs. Higher scores indicate better ability to recognize facial expressions.

In addition, the authors developed the Emotion Guessing Game to evaluate each child's ability to identify emotions portrayed by a live actor.

The cohort consisted of 71 children (mean age, 8.38 years; 63 boys [89%]). Of those children, 40 were randomly allocated to the treatment group, and 31 to the control group.

Sixty-eight of the children (96%) had been diagnosed with ASD using criteria from the Diagnostic and Statistical Manual, 5th edition (DSM-5). Three children (4%) had received a diagnosis of Asperger syndrome using DSM-IV criteria.

At baseline, all of the participants were receiving 15 to 20 hours of standard ABA therapy per week, and many were receiving additional interventions, such as speech or social skills therapy. They had similar baseline scores on the SRS-II, the VABS-II, and the NEPSY-II.

Overall, families in the treatment group "used the device on a mean of 12 times through the 6-week treatment period, 51% of the requested dosage of 24 days," including a mean of 3.9 sessions with the ABA behavior interventionist, which was 65% of the recommended number of sessions, the authors write.

Ultimately, 13 of the 40 children in the treatment group withdrew from the study; 27 completed the trial. Six of the 31 children in the control group withdrew before the end; 25 completed the trial.

Those children were offered the option of using the SG device for 6 weeks at the end of the study.

Limitations but First RCT to Show Benefit of Wearable Digital Device

In an intention-to-treat analysis, scores on the VABS-II socialization subscale increased significantly from baseline to the end of the intervention among children in the treatment group compared to those in the control group. The mean effect of treatment was 4.58 points (P = .005).

"The observed 4.58-point average gain on the VABS-II socialization subscale is comparable with gains observed with standard of care therapy," say the authors.

Positive mean treatment effects were also found for the other three primary measures but not to a significance threshold of P = .0125.

The findings "support the hypothesis that the SG intervention can improve social skills of children with ASD between the ages of 6 and 12 years as an augmentation to standard of care therapy," they write.

Statistically, however, the study was underpowered by a factor of 2, they note.

"Further limitations include low participant adherence to the recommended treatment dosing," restricting recruitment to families living within driving distance of Stanford University, "where the population may be enriched for familiarity with technology," and heterogeneity of social skills among children who entered the study.

In addition, the authors were unable to perform direct comparisons between the control and treatment groups after the end of the study, so information on sustained gains is lacking.

Despite these limitations, they conclude, "To our knowledge, this is the first randomized clinical trial to demonstrate efficacy of a wearable digital intervention to improve social behavior of children with ASD." They note that the findings "underscore the potential of digital home therapy to augment the current standard of care."

The study was supported by Google Inc, the National Institutes of Health, the Lucile Packard Foundation for Children's Health, the Hartwell Foundation, the Wallace H. Coulter Foundation, the Stanford Precision Health and Integrated Diagnostics Center, the Stanford Predictives and Diagnostics Accelerator, the Stanford Bio-X Center Program, the Stanford Beckman Center, the Walter V. and Idun Berry Postdoctoral Fellowship Program, the Bill and Melinda Gates Foundation, Peter Sullivan, and Bobby Dekeyser. The authors' relevant financial relationships are listed in the original article.

JAMA Pediatrics. Published online March 25, 2019. Full text


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: