In treating children with autism, it is critical in many programs and services to track significant amounts of data, both for reporting purposes and in assessing learning, behavior and intervention effectiveness. Current data collection techniques can be
extremely time consuming, may inhibit the caregiver-child interaction, may be subjective, and do not always communicate all of the information needed to fully assess the skills and treatment objectives under consideration.
In an effort to streamline data-gathering processes and support the therapy teams working with individuals with autism, Gregory Abowd, D.Phil. and his research team have received a CAN Innovative Technology for Autism grant award. With this grant, the team will build upon their prior work in testing the application of various emerging computing technologies in therapeutic and real-life settings. As these new technologies are further explored and developed, the hope is that they will allow clinicians the opportunity to collect more reliable data to track patterns in behavior and skill development, as well as promote better-controlled scientific studies of intervention effects. In addition, since the child is the one constant across a variety of settings and staff, the team is focusing on technologies designed to travel with the child so as to provide all caregivers access to relevant information and promote greater communication among providers. In practice, this might allow an afternoon therapist detailed information from a child's morning session, so that interventions might be targeted more specifically for the child's needs on that particular day.
Taking advantage of the inherent structured interactions and established collection routines of discrete trial training (DTT), one technology being explored captures DTT data and synchronizes digitally-stored therapist notations of trial performance (using a digital pen) with digital video of the interaction. The goal for this system is to improve data analysis, as well as link trial data to corresponding segments of video recordings so that researchers or lead therapists can readily search for specific interactions based on the recorded data. The Georgia Tech team is also investigating the possibility of reducing or eliminating the need for a therapist's manual recordings by exploring the use of automated capture technology that electronically tracks performance data using pre-programmed hand signals or other coded movements within the video. In both models, the availability of digital data and multi-modal presentations of information will allow for heightened insight into trends in performance, monitoring of consistency across therapists, and automatic tabulation of data and performance measures, therefore saving time and providing greater oversight opportunities. In the first test of this technology (being presented at a conference this fall), the team noted that the ability to readily search for specific video clips has proven to be a successful tool for use in DTT team meetings, and that the automated data tabulation resulted in a greater proportion of therapy time spent directly interacting with the child.
Significantly more challenging is the tracking of children's behavior in less structured interventions or in the natural environment, where additional data might provide insight into positive or negative trends reflecting the effectiveness of a variety of therapies (such as dietary, medical, or sensory-based interventions) or generalization of skills. Dr. Abowd is investigating using wireless technology and simple, portable data collection devices (approximately the size of a deck of cards) that can travel with the child across settings. In addition, applying state-of-the-art computer sensing can allow more streamlined interventions (data collection becomes automated and less intrusive), as well as expand the possibilities for collection of typically overlooked information. The team will explore the use of wearable sensors that can collect physiological measures (for example, accelerometers to detect and characterize the child's movements, such as 'stimming') or environmental factors (lighting, room temperature, etc.). These sensors can supplement the manual data collection of caregivers, allowing for the analysis of relationships between a child's behavior and important variables that may otherwise go unrecognized.
The team is also exploring the use of continuous running video cameras that only store information when triggered to do so-for example, a parent or teacher might press the record button just after a significant interaction or temper tantrum, saving only that episode for later review and evaluation. Multiple cameras or other sensors could be programmed to all respond to the same trigger, thereby collecting a broad array of environmental/situational data for further evaluation. The use of this technology in streamlining Functional Behavior Assessments is an area the team intends to explore.
Through this grant, as well as by providing access to related research and collaborative partners, Cure Autism Now is proud to support the Abowd team's efforts in applying technology to the challenges of data collection and analysis in the field of autism. For any therapist or parent who has followed a child around with a clipboard or compiled data in the wee hours of the night, technological advances in this arena cannot come soon enough!
Visit the ITA page in our Research Initiatives area of the Web site.