Towards a multimodal approach for assessing ADHD hyperactivity behaviors
Overview
This research investigates how sensor-based technologies can support more accurate assessments of ADHD, particularly hyperactivity in young children. Traditional evaluations often take place in clinical settings, which may not reflect how children behave in everyday environments. By reviewing existing clinical assessment criteria and aligning them with the capabilities of ambient and wearable tech, this study proposes a multimodal approach to collect contextual behavioral data. Design sessions with personas and scenarios further explore how these tools could enhance expert decision-making in real-world settings.
Methods
Prioritized accessibility, equity, and context-awareness in all phases of design, addressing diagnostic bias and promoting inclusive user experiences.
Conducted a structured behavioral analysis using ADHD assessment tools (DSM-V, Conners, SWAN) to extract and categorize measurable traits.
Mapped behaviors to sensor technologies such as wearables, motion tracking cameras, and microphone-based audio analysis, identifying opportunities for contextual data collection.
Explored digital health interventions and data capture methods using wearables, audio analysis, and gesture recognition.
Led design ideation sessions using Miro for affinity diagramming, brainstorming, and scenario development.
Created user personas and diagnostic journey scenarios to envision practical use of Ambient Intelligence (AmI) in real-world settings.
Contributed to inclusive, technology-driven, and evidence-based practices to support ADHD diagnosis in educational and clinical environments.