Machine Learning-Based Detection of Meltdowns in Autism Spectrum Disorder Individuals Using Galvanic Skin Response Signals

Abstract

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by challenges in social interaction, communication, and repetitive or restrictive behaviors. Individuals with ASD may have a range of symptoms and severity, which can include difficulty understanding and responding to social cues, sensitivity to sensory input, and a strong preference for routines. A meltdown is a common response for individuals with ASD when they are overwhelmed by sensory input, emotions, or changes in their environment. During a meltdown, the individual may lose control over their behavior, resulting in intense expressions of distress such as crying, screaming, or physical outbursts. Deep pressure therapy, such as that provided by a firm hug or weighted jacket, can be very effective in helping individuals with ASD   manage meltdowns. In this research, I proposed a machine learning-based meltdown detection from galvanic skin response signals. The proposed system automatically detects meltdowns in individuals by analyzing galvanic skin response signals obtained from a wearable device. The proposed system achieved an accuracy of 86.6% which demonstrates its feasibility.