A long-term goal of Guillermo Sapiro’s laboratory is to develop and validate sensitive, objective, reliable and scalable digital tools for measuring outcomes in autism spectrum disorder (ASD) treatment research. Behavioral observation and coding is still the gold standard in screening, diagnosis and outcome assessment for ASD. Yet behavioral coding is subjective, needs significant rater training, does not capture data from the participants in their natural environments and is not scalable for use in large populations or for longitudinal monitoring — all aspects that are critical for the success of multisite behavioral and pharmacological clinical trials.
To address this need, Sapiro — in collaboration with Geraldine Dawson at Duke University — seeks to develop and validate automatic, objective and quantitative measurements of social communicative and motor behaviors based on visual attention, affective facial expressions, vocalizations and head movements that can serve as feasible, reliable and sensitive outcome measures in ASD clinical trials.
Sapiro and Dawson have been developing automatic digital measurement tools that can reliably capture quantitative objective data from participants of a wide age range and functioning level in the laboratory, clinic or their natural environments, without the need for rater training or costly equipment1. The tools are based on computer video analysis that automatically codes a participant’s behavior while they watch stimuli carefully designed to elicit behaviors relevant to ASD symptoms. The stimuli, video recording and automatic analysis are all integrated in ubiquitous devices such as computers, phones and tablets, producing a software-only solution without need for specialized equipment/hardware, allowing the assessment of participants outside of a laboratory setting. The novel assessment framework, SenseToKnow, is based on active closed-loop sensing, where children are shown brief, developmentally appropriate, dynamic stimuli on a smart tablet, while the sensors in the same device capture information for automatic, objective quantification of several behavioral risk markers, based on patterns of attention, orienting, affect, vocalizations and motor behavior.
Sapiro was previously awarded a SFARI Explorer Award to validate these tools in a large population of infants and young children with ASD. In the current project, Sapiro and Dawson plan to extend this work by evaluating the psychometric properties of an adapted version of the framework that is developmentally appropriate for older children (preschool-school age), leveraging two studies (one is a case-control study; the other is a clinical trial) that are currently funded as part of the National Institutes of Health Autism Center of Excellence (ACE) at Duke University, led by Sapiro and Dawson. The specific aims are to evaluate the test-retest reliability of SenseToKnow, its construct validity, its concurrent validity (ASD, attention deficit hyperactivity disorder [ADHD], ASD+ADHD, and typically developing children) and finally to evaluate sensitivity to change in the context of a pharmacological clinical trial.
Although this project will focus on children in the preschool-elementary school age, the framework will be applicable to a wide age range by adapting the content of the stimuli. Furthermore, the approach is suitable for individuals of a wide range of functioning level, is scalable for multisite trials with up to 1,000 participants, presents an acceptable burden for participants and families, and can be used in naturalistic environments, including the home.