Predicting Situational Onset of Aggression in Minimally Verbal Youth with Autism Using Biosensor Data and Machine Learning Algorithms
Abstract
Unpredictable and potentially dangerous aggressive behavior by youth with autism spectrum disorder (ASD) isolates them from important educational, social, and family activities, thereby increasing the difficulties and costs associated with the condition. As many as two thirds of youth with ASD display aggression, which is one of the primary reasons they get referred for treatment. Aggression presents serious safety risks for the individual and others in the environment and frequently occurs with agitation, meltdowns, and other problem behaviors that are difficult to manage. Families report that aggression increases their stress, isolation, and financial burden and decreases available support options. Aggression to others is particularly impairing and difficult to manage in the 30%-40% of youth with ASD who are minimally verbal (MV-ASD). Their difficulty verbalizing distress can lead to behaviors that seem to occur without warning, sometimes long after any obvious trigger. This unpredictability makes aggression to others in MV-ASD dangerous and a barrier to accessing the community. This predicament can demoralize caregivers, accelerate negative trajectories, decrease quality of life, and collectively increase health care costs. Evidence-based pharmacological and behavioral interventions for aggression in ASD are frequently ineffective due to significant medication side effects or insufficient time to provide de-escalation strategies. Aggression to others may represent a maladaptive attempt to express or modulate physiological arousal arising from distress. Thus, we hypothesize that physiological arousal precedes aggressive behavior. Our project aims to predict aggression to others in MV-ASD before it occurs using data collected from commercially available wrist-worn wireless physiological sensors. The unique inpatient setting where this study will take place allows us to study aggression in a controlled, safe environment with 24-hour access to patients for an average of 3 weeks each. Data obtained in this setting thus reflects real-world biology and behavior in a safe environment. We also take advantage of technological advances with our combination of wearable physiological sensors, a smartphone application to document behavior, advanced digital signal processing, and sophisticated machine learning algorithms. This innovative approach has the potential to improve our ability to identify escalating distress in youth with MV-ASD, overcoming their inherent difficulty conveying feelings and emotions. By linking observable aggressive behavior to detection of preceding physiological signals (e.g., heart rate, skin temperature, sweating), we move the field of problem behavior assessment and treatment in autism toward a new biologically based, data-informed approach that is focused on prospective monitoring, prevention, and eventual real-time intervention. This project seeks to fill critical gaps in knowledge identified in several Department of Defense (DOD) and National Institute of Health (NIH) autism priority areas. First, it is responsive to the current priorities of the DOD Autism Research Program (ARP) to enable better understanding of factors that explain underlying conditions co-occurring with ASD, such as aggression. Second, as highlighted in the Fiscal Year 2017 ARP Idea Development Award announcement, our proposed project represents a new paradigm that challenges existing paradigms by incorporating state-of-the-art technologies and data analytics by a multi-disciplinary clinical, research, and engineering team to shift assessment and treatment of challenging behaviors to a biologically based approach. Third, the NIH Interagency Autism Coordinating Committee (IACC) Strategic Plan 2013 update calls for more research on the MV-ASD population, who are among the most difficult to treat of those diagnosed with ASD. Fourth, aggression has recently been highlighted as one of the most critical areas in need
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Oct 29, 2018
- Source ID
- W81XWH1810459
Entities
People
- Matthew Siegel
Organizations
- Maine Medical Center
- United States Army