Co-designing meaningful extended reality for physical rehabilitation: a stakeholder-driven approach to embodied telehealth.

Autism Spectrum Condition (ASC) is classified as a neurodevelopmental disorder in the Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5) []. In the UK, it is estimated that more than one in every 100 people is Autistic, with at least 700,000 Autistic children and adults nationwide []. Early signs of ASC often include delayed language development, repetitive behavioural patterns and limited or no social engagement. Our research focuses on developing toolsets and platforms that use camera data and machine learning to assess classroom engagement and physiological or affective states, such as arousal and dysregulation, in children with ASC. Given the heterogeneous nature of ASC, traits and behaviours associated with it exhibit very different patterns across individuals. Therefore, recognising and responding to children's emotional and behavioural cues in a classroom setting can be challenging. This challenge highlights the need to develop a multimodal system that monitors emotional dysregulation in students with ASC and alerts relevant caregivers or teaching staff to mitigate triggers for such events and prevent further escalation of the situation [].