SBIR Phase II: Using Data Mining to Optimally Customize Therapy for Individuals with Autism

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project will revolutionize the treatment of individuals with autism. One of every sixty-eight US children has autism (over 1.1 million). The estimated cost of providing Applied Behavior Analysis (ABA) therapy to those who could benefit is $7.5 billion dollars annually. Societal impacts include: 1) more individuals with autism across the globe will receive treatment regimens that will enable them to live more fulfilled lives and reach their full potential; 2) families whose children are good candidates for treatment and receive it will experience reduced stress and better family life; and 3) the additional lifetime cost of not effectively treating children with autism, which is approximately ten-fold the cost of treatment, will be reduced. Because high-quality, contextually rich ABA performance data will be collected for the first time, efforts to apply data analytics will contribute in two important ways: a) patterns may be discerned across individuals with autism to better understand variations in autism and create therapies to target these differences; b) expansion of the frontiers of data mining to provide guidance in real time will contribute to a number of areas within and beyond ABA therapy. The proposed project will optimize therapy outcomes for individuals with autism by transforming agent-based guiding technology into an adaptive and intelligent ABA therapy assistant for supervisors and instructors. The project pushes the boundaries in providing cost-effective, adaptable, intelligent, real-time guidance and data-collection support to instructors that integrates naturally into the instructional process and is easy to learn and use. ABA therapy experts, supervisors and instructors will verify the analyses and resulting guidance incorporated into the technology. Advanced theories of usability engineering, including some developed by the project team, will be used to build interfaces that supervisors and instructors can intuit without the need for learning new concepts and syntax. The project will utilize the collected logs from multiple sessions with multiple therapy recipients and multiple therapy providers to uncover hidden patterns and assist supervisors in selecting appropriate therapy steps personalized for the individual with autism. The project will build on a large body of recent work in visualization, machine learning on temporal predictive modeling and sequential pattern mining, including some of the previous results of the project team. Special attention will be paid to the recent work in educational data mining and intelligent tutoring.