Early and Accessible ADHD Screening: A Design Proposal Integrating the Digital EDAH Test with Computer Vision

Calla J.S.R.; Suarez D.J.N.; Silva P.D.R. · Proceedings - 2025 5th International Conference on Information Systems and Software Technologies, ICI2ST 2025 (2025) · DOI: 10.1109/ici2st68465.2025.00027

Resumen

Attention Deficit Hyperactivity Disorder (ADHD) is one of the most common neurodevelopmental disorders in childhood, significantly affecting school performance, emotional self-regulation, and social interaction. Despite its high prevalence, many educational institutions lack accessible tools for early detection. This paper presents a solution proposal based on a web application that integrates an adapted digital behavioral test (EDAH) with a computer vision system for automated analysis of attentional variables. The architecture includes five key components: a child-friendly web interface, a structured backend layer, an intelligent diagnostic module, a relational storage system, and a camera-based visual analysis subsystem. As an initial feasibility test, a set of machine learning models (Random Forest, XGBoost, SVM) will be trained on a balanced dataset, with the expectation of achieving promising performance goals under controlled conditions. These performance expectations, although encouraging, represent an idealized baseline and should be validated in real contexts with more diverse and unbalanced data. The proposal demonstrates potential as a scalable and replicable alternative for school environments with limited resources, enabling broader access to reliable and non-invasive screening processes. Lines of future work are proposed to clinically validate the system and strengthen the capabilities of the visual module.

Instituciones: Universidad Peruana de Ciencias Aplicadas